Sie sind auf Seite 1von 74

GoldWave Manual

Copyright 2002, 2005 GoldWave Inc.

Do not upload or include this document on a website. March 2002

Table of Contents
I. Introduction 1. Features How to Use This Manual 2. 3. Notation Getting Started 1. System Requirements 2. Installation Installation From a Downloaded Program (Exe) File a. b. Installation From a Downloaded Zip File c. General Installation Instructions i. Adding a Shortcut ii. Setting Audio Devices iii. Additional Settings Using GoldWave 1. Interface Overview 2. Device Controls Overview a. Properties i. Playback Properties ii. Record Properties iii. Volume Properties iv. Graph Properties v. Device Properties b. Playing Sounds c. Pausing Playback d. Stopping Playback e. Rewinding and Fast Forwarding f. Recording Sounds g. Volume and Balance Faders h. Speed Fader 3. Editing Overview a. Selecting Part of a Sound b. Editing the Waveform with the Mouse c. Mixing and Cross-Fading 4. Storage Overview a. Direct-to-Disk b. RAM

II.

III.

5.

6.

7.

c. Flash File Menu Commands a. File Format b. New c. Open d. Close e. Information f. Batch conversion g. Save h. Save as i. Save selection as j. Exit k. File History Edit Menu Commands a. Undo b. Cut c. Copy d. Copy to e. Paste new f. Paste and Paste at g. Mix Replace h. i. Delete j. Trim k. Insert silence l. Select view Select all m. n. Channel o. Marker i. Set ii. Recall positions iii. Store positions iv. Drop start/finish v. Snap to zero-crossing p. Deflash Effects Menu Commands a. Special Controls for Effects i. Presets ii. Shape Controls b. Doppler c. Dynamics d. Echo e. Expand/Compress f. Filter i. Noise gate ii. Noise reduction iii. Low/Highpass iv. Bandpass/stop v. Equalizer vi. Parametric EQ vii. Pop/Click viii. User defined g. Flange

IV.

Interpolate Invert Mechanize Offset Pitch Reverse Silence Stereo i. Exchange channels ii. Pan iii. Remove vocals p. Time warp q. Volume i. Change ii. Fade in iii. Fade out iv. Maximize (Normalize) v. Shape r. Playback rate s. Resample 8. View Menu Commands All a. b. Other c. Previous zoom d. Selection e. User f. Zoom in g. Zoom out h. Zoom 10:1 and 5:1 i. Zoom 1:1 j. Zoom 1:10, 1:100, 1:1000 Vertical zoom all k. l. Vertical zoom in m. Vertical zoom out n. Start and Finish 9. Tools Menu Commands a. Cue points b. Expression evaluator c. CD player d. Volume controls e. Device controls f. CD audio extraction 10. Options Menu Commands a. Colours b. Controls c. File d. File types e. Tool bar f. Window 11. Window Menu Commands 12. Help Menu Commands General Information 1. Warranty, Trademarks, and Copyright

h. i. j. k. l. m. n. o.

2.

Support and Updates

A. Appendix: An Introduction to Digital Audio 1. Digital Audio Attributes 2. Byte Swapping and Sign 3. Frequency and Pitch a. Frequency Ranges b. Frequency Range and Sampling Rate c. Frequency Spectrum Graphs 4. Problems with Recording B. Appendix: Keyboard Commands C. Appendix: Expression Evaluator 1. Overview 2. Evaluation Range, Variables, and Constants a. Using Time in an Expression b. Using the Sample Index in an Expression c. User Variable f d. Conversion Between Variables 3. Group name and Expression name 4. Evaluator Operators and Functions 5. Signal Generation 6. Custom Filters D. Appendix: Troubleshooting and Q&A 1. Troubleshooting 2. Common Questions and Answers

1 Introduction
GoldWave is a comprehensive digital audio editor that plays, records, edits, and converts audio on your computer. This section lists some of the features of GoldWave and outlines the notation and organization of the manual.

Features
GoldWave includes a full set of features that rival even the most expensive audio processors:

An intuitive and customizable user interface makes editing easy. An independent Device Controls window provides direct access to audio devices. It contains controls for playback, rewind & fast forward, recording, volume, and speed. LED meters and real-time graphs display the sound during playback and recording. A multiple document interface (MDI) allows many files to be opened at one time, simplifying file-to-file editing. Huge files can be edited using hard disk storage. Small files can be edited quickly using RAM storage. Sounds are displayed graphically as a waveform and the level of detail can be changed by zooming in or out. The waveform can be reshaped directly with the mouse when zoomed in. Many audio effects, such as Reverse, Echo, Pan, Flange, Dynamics, and Time warp, enhance, distort, or alter sounds in various ways. Sophisticated filters such as the Noise reduction and Pop/Click filters help restore audio. The Batch conversion command converts a group of sound files to a different format and type. It can convert stereo to mono, 8 to 16 bits, or any combination of attributes supported by a file type. With an MPEG codec installed, you can compress audio to a fraction of its original size while still keeping excellent quality in MP3 format. The CD Audio Extraction tool copies audio from a CD to a file on your system. Audio can be extracted directly to MP3 format to minimize storage space. An Expression Evaluator allows you to generate everything from simple tones to complex filters. Expressions for telephone dial tones, waves, and effects are included.

How to Use This Manual


Familiarity with the Windows interface, such as property dialogs, tool bars, etc., is recommended before reading this manual. For those who are unfamiliar with digital audio, Appendix A briefly introduces some of the fundamentals of computer audio. It also provides some solutions to common recording problems. Appendix D contains troubleshooting information and answers to common questions. Section 2: Getting Started, covers system requirements and installation. Section 3: Using GoldWave explains the interface and menu structure in detail. Topics are covered in the order that they appear in GoldWave's menu. Section 4: General Information, provides support, copyright, and warranty information.

Notation
Bold text and a vertical bar are used to denote menu commands. File | New, for example, means to select the New command from the File menu. This notation is used to refer to other sections within this manual as well. In the above example, you can find information by looking for New under the File Menu Commands section. If the first word is Start, select the command from the Windows task bar menu structure. A pointing hand emphasizes helpful information and techniques. An exclamation mark emphasizes warnings and other important information.

2 Getting Started
System Requirements
The minimum system requirements for GoldWave are:

Pentium based PC or compatible Microsoft Windows 95 or Windows NT 4.0 or higher 32 Megabytes of RAM (64 MB recommended) 2 Megabytes of hard disk space Mouse Sound card with a Windows driver

If you need to edit large files, you will need a large amount of hard drive space. One minute of CD-quality sound requires 10 megabytes of storage. GoldWave may require 20 to 40 megabytes per minute when editing an existing file and the Undo feature is enabled. For editing audio in movie files and editing mp3 files, you must have the new DirectX Media extensions installed. You can install the new Media Player available from Microsoft's website to get these extensions.

Installation
The following two sections give instructions for installing GoldWave on your system. Before running GoldWave make sure that you have an appropriate Windows 95 sound driver installed. If you need to add one, use the "Add New Hardware" item under Start | Settings | Control Panel. The driver and installation instructions should be included with your sound card. The current settings for your sound card are listed under the "Sound, video, and game controllers" item of the Device Manager. The Device Manager is found under "My Computer" Properties or the System icon in the Control Panel.

Installation From a Downloaded Program (Exe) File


If you downloaded the self-installing version of GoldWave, simply run the program. You can specify a destination folder where GoldWave will be installed. A desktop shortcut and Start menu entries can be created automatically.

Installation From a Downloaded Zip File


To install GoldWave from a zip file, you will need PKUNZIP version 2.04g (or compatible): http://www.pkware.com Create a new folder called "GoldWave", or select your current GoldWave folder and unzip the GoldWave zip file into it. Follow the general instructions below to complete installation.

General Installation Instructions


Check the readme.txt file for any additional information not available at the time this manual was created. New versions of GoldWave can be downloaded from the web site: http://www.goldwave.com

Adding a Shortcut
To add a GoldWave icon to your desktop, right click the mouse pointer on an unused area of your desktop to display the context menu and select New | Shortcut. Choose Browse, find and double click on goldwave.exe, then click on Next. Type in "GoldWave" for the name of the shortcut, then click Finish. To run GoldWave, either double click on the new GoldWave icon or use the Start | Run command.

Setting Audio Devices


GoldWave allows you to choose which devices to use for playback and recording. Choose the properties button on GoldWave's Device Controls window, then choose the Device tab. Drop down lists of installed playback and recording devices and their capabilities are presented. Select appropriate devices from the lists. Note that if your device only supports 8 bit recording, you must select the Microsoft Sound Mapper device so that GoldWave can record.

Additional Settings

Use Options | File to set your sound folder. Use Options | Tool bar to customize tool bars. Use Options | Colours to change Sound window colours.

To associate file types with GoldWave, such as wav or mp3 files: 1. 2. 3. 4. 5. Run GoldWave Choose File types from the Options menu. Enter the type name, such as wav. Choose the OS Associate button. Choose Close.

Now when you double-click a wav file, GoldWave starts automatically.

3 Using GoldWave
Interface Overview
GoldWave is composed of three windows: the Main window, the Device Controls window, and Sound windows. The Main window contains the main menu, two rows of tool bar buttons, and status bars (see Figure 1). It groups together and manages all the Sound windows.

Figure 1: Main Window The tool bar buttons provide quick access to many of the frequently used commands. The upper bar holds File, Edit, and View commands, while the lower bar contains Effects and Tools commands. The function of each button is displayed in the lower status bar when the mouse pointer is positioned directly over it. Tool bars can be configured using the Options | Tool bar command. The status bars show attributes of the Sound window, including the sampling rate, length, selected region, channels, and general file format information. By clicking the mouse pointer over any status item that shows time (shown as * in Figure 2), the unit or format for that status item can be changed. If you click the mouse pointer over the length item, for example, you will be presented with a menu showing length in terms of storage size, time, and samples.

Figure 2: Status Bar Contents Sound windows are created when you open a file. These windows contain a waveform graph of the sound with a time axis near the bottom. For stereo sounds, two separate graphs are shown.

The top green graph is the left channel and the bottom red graph is the right channel. The selected part of the sound is highlighted with a blue background. Near the bottom of the Sound window, a small "Overview" bar representing the entire sound shows you what part of the sound is selected (highlighted green and/or red), what part is displayed in the above graph (black background), and what part is not visible in the above graph (dark grey background). Initially, the entire sound is selected. You can change the selection by using the left and right mouse buttons. You can configure the window size and axes format of Sound windows using the Options | Window command. The Options | Colours command sets the colour scheme. The Device Controls window interacts directly with your sound card. It contains buttons to play and record sounds as well as controls for volume, balance, and playback speed (provided your audio device supports these features). LED meters and real-time graphs display audio data whenever a sound is played or recorded. Detailed information about the graphical displays and controls is presented in the next section.

Device Controls Overview


The Device Controls window (see Figure 3) is an easy-to-use interface to your audio hardware and drivers. On the bottom half of this window are graphs in which sound is displayed during playback and recording. On the top left section of the window is a standard set of audio controls, including play, stop, record, rewind, pause, and fast forward. LED meters are located below these controls. In the top right section of the window are controls to set the device's output volume, balance, and playback speed. The Device Controls window can be resized to change the size of the graphs or to hide them.

Figure 3: Device Controls Window

Properties
The Properties button presents a property sheet containing several pages to configure playback, recording, volumes, graphs, and devices. These options are described in the following sections. After installing GoldWave, you should take a moment to see if the correct playback and recording devices are selected under the Device tab and familiarize yourself with the settings under the Playback (Figure 4) and Record tabs.

Figure 4: Device Controls Properties

Playback Properties
The Playback Properties page contains options to configure the User play speed for rewind and fast forward. The User play options are as follows: Option All Selection Unselected Button function Plays entire sound. Plays region between start and finish markers. Plays regions outside the start and finish markers. This lets you quickly test how a cut or delete will sound without actually changing the sound. When possible, playback is confined to the region shown in the Sound window view so that the entire sound does not have to be played. Plays all of the sound currently shown in the Sound window view. This is useful if you zoomed in on part of the sound. Plays three seconds just before the finish marker, so you can determine if that marker is in the right place without listening to the entire selection. This is a special playback feature that plays the sound in three sections. The beginning of the sound, outside the selection, is played first. Then the selection is played and looped. Finally the end of the sound, outside the selection, is played. This is useful for musical accompaniment or looped instrument samples. If checked, it specifies the number of times playback should be repeated. A value of 1 loops playback once, so the selection is played twice. A zero value loops forever. button and set the

View Finish

Intro/loop/end

Loop

Fast/Rewind speed The playback speed of the fast forward and rewind buttons is controlled by these values. A value of 1.00 is normal speed. Entering a value of 3.00 for Rewind speed, for example, means the rewind button will play the sound backwards three times faster than normal. By entering small numbers (such as 0.1) the rewind and fast forward buttons will play very slowly. This is useful for finding pops or clicks, since the graphs will move slowly through the data.

Record Properties
The Record Properties page includes features to monitor the recording sources, start recording automatically when a sound is detected, or delay recording for a certain length of time. The basic options are as follows: Option Monitor Purpose Connects the recording source to the graphs and LED meters so you can adjust volume levels before recording. See Recording Sounds for information about selecting a different recording source and setting volumes. Restarts recording when the end is reached and continues to record over and over until the stop button is pressed. This is useful if you are trying to capture a sound but do not know when it might occur. By loop recording a 1 minute sound, you will always have the last minute of audio stored for recall. Prevents you from accidentally recording over a sound. To record, you must hold down the Ctrl key, otherwise a safety message appears. Saves the entire selection so that you can undo recording after. If you record large files, you should not check this setting. Mixes the newly recorded sound with the existing sound. This lets you layer recordings. Immediately writes the newly recorded sound directly to the hard drive. Normally, Windows will temporarily save the sound in memory (cache) before writing it to the hard drive. However, when recording large files, the memory quickly becomes full and Windows is forced to write a large amount of data. This can cause skipping and gaps in the recording. In some cases, disabling write cache can solve this problem. Continues recording until all storage is exhausted or until you press the record stop button. The file size is increased automatically to hold the new audio, which can save time when recording for a long period of time. You must enable Hard disk storage under Options | File when using this option. If you are using RAM storage, this option is ignored.

Loop

Ctrl key safety

Allow undo Mix with selection Disable write cache

Unbounded

Delayed recording The Timer delays recording until the specified time and day of the week. Use this feature to automatically record something at a later time. The time is given in 24 hour time. A time of 06:00:00 is 6:00 AM and a time of 18:00:00 is 6:00 PM. 00:30:00 is 12:30 AM or 30 minutes past midnight. If entering the time directly, remember to include the seconds. Entering 18:00 means 00:18:00. You must press the record button to activate the timer. Level activated recording is useful for automatically synchronizing recording to a sound source or efficiently capturing airport or police radio where there may be a lot of silence that does not need to be recorded. It automatically starts recording when the sound source is above a given level and pauses recording when the sound is below the level. The Threshold specifies how loud a sound should be before recording begins. A value less than 0.2 is typical. The Duration specifies how long to record after the sound becomes quiet again. Using a value of 3 allows recording to continue for three seconds after the sound goes below the specified threshold. To minimize silence, use a value of 1 second or less, but not zero. A zero value causes recording to continue without stopping once the level first activates. The Time stamp cues option provides a way of marking the date, time, and position of each activation. Cue points with the current date and time are created and can be view under the Cue points tool.

Volume Properties
The Volume Properties page lets you adjust recording volumes and select or unselect recording sources. A volume fader and checkbox is shown for each source. To select a source, check the appropriate checkbox. If your sound card has a master control, make sure that the Mute all option is turned off and that the master volume is not zero. You can use the Monitor option under the Record Properties tab to activate the LED meters and real-time graphs without recording. By moving the Properties window, you can see the levels as you adjust the volume faders. Note that volumes are changed regardless of whether you choose OK or Cancel. Using Cancel will prevent the left and right graphs from being set the same when you close the Properties window. To select a different recording device, use the Device Properties page.

Graph Properties
The Graph Properties page controls the graphs and LED meters. The graphs display audio data in a variety of ways:

Graph Amplitude Spectrum Spectrogram

Description Standard amplitude waveform. Frequency analysis of the sound. Coloured frequency spectrum, with time on the x-axis, frequency on the y-axis and colour as the magnitude. The colours, in increasing magnitude, are black, purple, blue, cyan, green, yellow, red, and white. A cyan point, for example, is higher magnitude than a blue point. Detailed coloured bar graph, similar to the spectrogram colours given above. Detailed fire coloured bar graph. Logarithmic frequency band bar graph commonly found on stereo systems. The sound is plotted with the left channel against the right channel to generate a Lissajous pattern. This is often used to see the phase difference between two equal frequency signals. If the left and right channels are in phase, the pattern is a diagonal line running from the lower left to the upper right. If the channels are 90 degrees out of phase, the pattern is a circle. For general stereo sounds, it looks like a crazy scribble. The larger the scribble, the larger the difference between the channels. Monaural sounds always show a diagonal line since the left and right data are the same. As above, but coloured points are used instead of lines. This is the same as the Amplitude graph above, but coloured points are used instead of lines.

Colour spectrum Fire spectrum Log bar spectrum X-Y mode

Colour X-Y mode Colour Amplitude

The axes of these graphs will be numerically labelled if you check the Show axis box. Note that you may have to resize the Device Controls window to make more room. Window function When calculating any frequency related graphs, a window function must be used to smooth out analysis. The Kaiser 7 or Hamming windows are usually the best, but you can try the other windows just to compare the results.

Refresh Frames/s sets the number of times per second that graphs and LED meters are updated. A value of 30 or less gives good results, but you may want to use 60 to get an extra detailed spectrogram. The LED hold time is the amount of time, in milliseconds, that the peak LED meter level is retained. Note that the top red overload LED remains lit until the sound is stopped. Fade time is the amount of time, in milliseconds, that it takes for the LED meters and the colour and fire spectrum graphs to drop from maximum to zero.

Device Properties
The Device Properties page lists playback and recording devices and shows the capabilities of the selected device. Normally, you do not need to change these settings unless you have more than one sound card or you encounter a problem. Some sound card drivers do not report the current playback/record position correctly. In such cases, you will notice that the time in the status bar is incorrect and the white "current position" line in the Sound window is moving too fast or too slow. If this happens, select a different Positioning setting. If you notice memory errors or gaps during playback or recording, change the Buffer size settings. The Buffer size determines the amount of audio, in seconds, that is transferred from the device driver to GoldWave. Values of 1 second or less give the best results. If you get a "capabilities" error, select the Microsoft Sound Mapper playback or recording device from the drop down lists. The Joystick control option allows you to control playback and recording using a standard joystick. The first available and attached joystick is used. The directions control playback with left for rewind, right for fast forward, down for pause, and up to unpause. The first button (button 1 or A) starts and stops playback. The second button (button 2 or B) starts and stops recording.

Playing Sounds
After opening a sound (see File | Open), you can use the play button or the User play button to play it. The play button plays the selection only. The User play button can play or loop the entire sound or certain parts of it. This button is configured by the Playback Properties page. While a sound is playing, it is displayed on the real-time graphs and LED meters. The current position is displayed in the graph of the Sound window as a white, vertical line. You can move the start and finish selection markers to the playback position by using the bracket keys, [ and ] or Edit | Marker | Drop... commands. See Editing Overview for more information about changing the selection. To play the entire sound, configure the User play button to All.

Pausing Playback
While a sound is playing, you can pause it with the pause button. Remember to use either play or stop later. Pause freezes the graphs and the current position marker so you can see the shape of the sound or move the selection markers.

Stopping Playback
Playback can be stopped immediately with the stop button. The current position is reset to the beginning. The audio device is released so that it may be used by other applications. Note that recording is stopped using a different button.

Rewinding and Fast Forwarding


You can use the rewind button or fast forward button to quickly move back and forward through the sound. The current position is displayed in the graph of the Sound window as a white, vertical line. You can adjust the speed of rewind and fast forward with the Playback Properties page, as described previously.

Recording Sounds
Use the record button to record your own sounds. Before you start, you need to create a new file using the File | New command. Audio is recorded into the selection of the Sound window replacing any audio that was previously there. You can make room for recording in an existing sound using the Edit | Insert space command. Recording stops automatically when the selection is full. You can stop recording at any time with the special stop button that appears in place of the record button. Many recording options are available in the Record Properties page. The Unbounded option allocates more room for recording automatically so that you do not need to set the length in advance. Sound cards usually have several recording sources, such as microphone, line-in, CD, and MIDI. To select and adjust a recording source, you need to use the Volume Properties page or the Volume Control accessory. To use the Volume Control accessory, use Start | Programs | Accessories | Multimedia | Volume Control or simply select Tools | Volume control in GoldWave. Next, use the Options | Properties command. In the properties dialog, select the Recording option and make sure all the items in the list are checked. Finally, choose OK to see the Recording Control. You can select a recording source and adjust the volume. To adjust the volume before recording, use the Monitor option under the Record Properties page. Remember to press the playback button on the CD player, if you are recording a CD. GoldWave records using 16 bit mode only. If you have an 8 bit card, you must select the Microsoft Sound Mapper device for recording.

Volume and Balance Faders


If your audio device supports volume control, you can use the volume fader to change the playback volume of your audio device. Move the thumb right or click the plus button to increase the volume. Move it left to decrease the volume. The current volume is shown numerically to the left of the fader. A value of 100 is full volume. If your audio device supports independent left and right volume control, you can use the balance fader to change the device's left/right balance. Move the thumb in the direction you want to shift the balance. Clicking on balance icon will quickly reset the balance to center. Note that these controls do not change the recording volume. See Recording Sounds for more information.

Speed Fader
The playback speed of the audio device can be changed with the speed fader. Move the thumb right to increase the speed, and left to decrease it. The relative speed is shown numerically to the left of the fader. Clicking on the speed icon resets the speed to normal. Note that changing the speed also changes the pitch like spinning a vinyl record faster or slower.

Editing Overview
Selecting Part of a Sound
Almost all commands in GoldWave operate on the currently selected part of a sound. The selected part, or selection, is the highlighted part of the sound between two vertical markers (see Figure 1). The vertical markers are cyan lines located to the left side (start marker) and right side (finish marker) of the view. Note that GoldWave does not use the standard "click-and-drag" method to make a selection because it does not allow accurate positioning of both markers. Instead, it uses an innovative method that lets you independently set the start and finish markers to the nearest sample (when zoomed in at a 1:1 level or better). The mouse pointer appears as moved.

when the markers can be

To move the start marker, click the left mouse button on the graph at the point where you want to move it. To move the finish marker, click the right mouse button on the graph at the point where you want to move it.

Additional notes and techniques:


You cannot place the finish marker before the start marker. The same is true for setting the start marker after the finish marker. You can set a marker's position by clicking in the Sound window's Overview box. The Edit | Marker | Snap to zero-crossing feature helps to minimize pops and click by finding a point where the waveform is close to zero amplitude. To find a certain part of the sound, you can play the sound and mark it as it plays using the [ and ] (bracket) keys. GoldWave indicates the part being played with a white, vertical line. You can use the rewind and fast forward buttons to pinpoint a sound. To select a single channel of a stereo sound use the Edit | Channel command or the Chan button. In most cases, editing and effects are performed only on the selection. Some effects, such as the Expression evaluator, Resample, and Playback rate alter the entire sound. The Windows clipboard is used for most of the editing commands. If the selection is too large to copy into the clipboard, GoldWave will automatically create a temporary file to simulate the clipboard. If you notice any freezes during editing, use the Options | File command to always use the GoldWave clipboard.

Editing the Waveform with the Mouse


You can directly edit the waveform with the mouse to remove pops and clicks or draw new sound waves. To do this, you must first zoom in so that individual samples are visible (see View | Zoom 5:1 or View | Zoom 10:1). If the sound was flash opened, you will need to use the Edit | Deflash command to prepare the sound for editing. 1. Choose Deflash from the Edit menu, if enabled. 2. Zoom in 2:1 or closer. 3. Place the mouse arrow directly over the waveform. The arrow will change into a thin horizontal line with two small arrows pointing to it. 4. Click and hold the left mouse button. 5. Move the mouse to change the waveform. 6. Release the mouse button to finish the changes.

Mixing and Cross-Fading


The Edit | Mix command mixes one sound with another so they both play at the same time. If you wanted to add vocals to music, for example, you would perform the following steps: 1. 2. 3. 4. 5. Open the sound containing the vocals. Choose Copy from the Edit menu. Open the sound containing the music. Choose Mix from the Edit menu. Enter a volume to mix the vocals (100 = full volume).

When mixing more than a couple of sounds, you should reduce the mixing volume and the destination volume to prevent clipping distortion. The volume of the destination sound can be reduced before mixing by using the Effects | Volume | Change command. To cross-fade two sounds, such as having one song fade out while another is fading in, follow these steps: 1. Open the first song. 2. Move the start marker to select the last 3 seconds of the song. 3. Choose Fade out from the Effects | Volume sub menu and use a 100% fade. 4. Open the second song. 5. Move the finish marker to select the first 3 seconds of the song. 6. Choose Fade in from the Effects | Volume sub menu and use an initial volume of 0%. 7. Choose Select all from the Edit menu. 8. Choose Copy from the Edit menu. 9. Click on the first song to activate that Sound window. 10. Choose Mix from the Edit menu.

Due to a problem with the Windows clipboard, Windows may freeze when copying a large file. If this happens, use the Options | File command to select the GoldWave clipboard option.

Storage Overview
GoldWave supports both direct-to-disk editing and RAM (physical computer memory) editing, along with a time saving flash feature. These features are configured using the Options | File command. If you have 32 megabytes of RAM or more and often edit files less than 10 megabytes in size, the RAM storage option gives the best performance.

Storage Overview

Direct-to-Disk
In direct-to-disk editing, the entire sound is stored in a temporary file where it can be modified. This allows you to edit very large files (up to about 1 billion bytes in size) provided the required disk space is available. Only a small amount of RAM is required for each opened sound. The drawback is that editing and effects processing take more time since audio data must be transferred to and from the disk.

RAM
In RAM editing, the entire sound is stored in memory. This allows you to edit and process files very quickly. It saves time and reduces the burden on your hard disk. The drawback is that the size of the files must be small enough to fit in the available RAM (not including virtual memory).

Flash
The flash feature opens large files instantly. The entire file is not copied to temporary storage and only a few seconds of the sound is initially graphed. This can save a great deal of time if you only want to play a file and not modify it. A flashed file can be deflashed at any time by using the Edit | Deflash command. In most cases, GoldWave automatically deflashes the file for you. A flashed file requires no extra disk space and only a small amount of RAM, which means that several large files can be opened at once, regardless of how much free space is available on the disk. The drawback is that you need a fast system when a file has to be decompressed for playback.

File Menu Commands


This section explains all the commands under the File menu. Several features for storing and handling files can be configured using the Options | File command.

File Format
Sound files come in a variety of forms. Usually, the form or type of sound can be determined from its filename extension, such as .wav or .mp3. GoldWave supports all the sound types listed in Table 1. Each file type can have several sub-formats or attributes. The .wav type for example, can hold audio encoded or compressed in dozens of different ways, including PCM, ADPCM, companded, or MPEG1 Layer 3. Table 1: Supported File Types Extension .au Comments Sun or NeXT files, commonly used on web pages and in Java. Supports 8 & 16 bit linear, mu-law and A-law encoded files. Any header block is preserved. Apple / Macintosh sound files. The blocks NAME, COPY, ANNO, AUTH, and CHAN are all preserved. Compressed files are not supported. Markers are not supported. Microsoft audio and/or video files. GoldWave can extract the audio portions of these files. The new Microsoft Media Player must be installed for GoldWave to open these files. DiamondWare sound files. Supports 8 & 16 bit PCM attributes. Amiga 8SVX files. The blocks NAME, COPY, ANNO, AUTH, and CHAN are all preserved. Matlab files. The data must be normalized (i.e. -1.0 to 1.0) for double precision data. If the "wavedata" variable is two dimensional, the data is assumed to be stereo. GoldWave saves audio data in the "wavedata" variable and the rate in the "samplingrate" variable. A 11025 Hz sampling rate is assumed if none is present. Quicktime movie files. GoldWave can extract the audio portion of the file (if present). The new Microsoft Media Player must be installed for GoldWave to open these files. MPEG1 Layer 3 compressed files. To read these files, you must have an MPEG codec installed. This codec is installed if you install the new Microsoft Media Player. To save a file in this format, you must have the LAME or BladeEnc encoder installed. See the GoldWave website for details. Ogg audio or video files. To read these files, you must have the Commom Audio Layer Ogg module installed (calogg??.dll). The module is available on the GoldWave website.

.aif .afc .asf .avi .dwd .iff .mat

.mov

.mp3

.ogg

.raw .sds .smp .snd .txt .wav

Headerless files containing binary data in 8 bit, 12 bit, 16 bit, double precision, mu-law, or A-law format. MIDI sample dump standard format. Loop points are not supported. Sample Vision 16 bit PCM sound files. Markers/Loops are not supported. Raw or NeXT files. NeXT files are automatically detected. Raw files present the File Format dialog for attributes. An ASCII text file containing a series of numbers. RIFF WAVE 8 & 16 bit PCM mono or stereo, A-law encoded, mulaw encoded, and Microsoft ACM compressed files. MPEG compressed audio is support only if the MPEG codec is installed. Only files with one 'data' chunk are supported. The chunks LIST INFO, LIST adtl, and 'cue' are detected. All others are ignored.

.voc

Sound Blaster files. Supports: 8 bit mono/stereo, 16 bit mono/stereo, mu-law encoded mono/stereo. ADPCM compressed files are not supported since the compression algorithm must be licensed from Creative Labs. Dialogic ADPCM encoded raw files. The File Format dialog is presented where you can specify the Telephony type and 4 bit VOX ADPCM format. You can use the Options | File types command to automatically select this format for all .vox files.

.vox

Normally, GoldWave detects and automatically opens all the supported file types. However, there are several cases where GoldWave may not be able to open a file: 1. The file type is not supported by GoldWave. 2. The file does not contain any header information and there is no extension association (see Options | File types). 3. The file contains compressed or encoded audio that GoldWave cannot decompress or decode. 4. The codec to decompress the file is not installed on your system. 5. The file type is recognized, but the file structure is invalid or corrupt. Unfortunately, there are many programs which create badly structured .wav files. If any of these conditions occur, GoldWave presents the File Format dialog (Figure 7) so that you can specify the format and attributes manually. If you have a file that contains audio copied directly from a CD, for example, you would choose the PCM format and 16-bit, stereo, signed attributes, with a sampling rate of 44100 Hz. The basic formats are given in Table 2.

Figure 7: File Format Table 2: File Format Description Format PCM Description and Attributes Audio is uncompressed 8, 12, 16, or 32 bit data. A Windows system usually creates 8 bit, unsigned or 16 bit, signed data. A Macintosh system usually creates 8 bit, signed and 16 bit, signed, byte swapped data. The signed attribute tells GoldWave how the bits should be interpreted. The byte swapped attribute tells GoldWave to change the order of bytes from big endian to little endian. Audio is in a compressed format used in telephone applications. This includes mu-law, A-law, ISDN A-law (inverted A-law), and 4 bit ADPCM VOX Dialogic files. Audio is binary IEEE floating point single precision (32 bit) or double precision (64 bit) data. The byte swapped attribute can be specified, but is usually not necessary. Audio is a plain text (ASCII) file containing numbers. The float attribute tells GoldWave that the numbers range from -1.0 to 1.0. The integer attribute tells GoldWave that the numbers range from -32768 to 32767.

Telephony

Floating point Text

If you do not know the format, experiment with trial-and-error. Start with an 8 bit or 16 bit PCM format, then try the mu-law or A-law Telephony formats. Generally, sounds will be noisy if the format or number of bits is incorrect, in which case you will have to close and reopen the sound using a different format. You can leave the sampling rate unchanged since it affects only the playback speed and can be changed later using Effects | Playback rate. Appendix A has more information about sound attributes.

New
Use New to create a new sound with attributes you specify. These attributes are discussed in Appendix A. Note that GoldWave allows you to create and edit sounds that may not be playable with your audio hardware. If you encounter any "capabilities" error messages, try selecting the Microsoft Sound Mapper device under Device Controls Device Properties. For Web and Java applications, you should specify mono, with a sampling rate of 8000 Hz. For CD quality, use stereo, with a sampling rate of 44100 Hz, or simply choose the CD button.

Open
The Open command presents a list of files in your sound folder. The default sound folder can be set using the Options | File command. All file types having a recognized extension are listed. After you select a file, a Sound window is opened and details about the sound are displayed in the status bar. See the File Format section above if GoldWave could not open the file. The Storage Overview section explains how the sounds are stored for editing. Depending on the size of the file, you may want to change the storage setting under Options | File.

Close
Use Close to close the current sound. If any changes were made, you will be asked to save them.

Information
Sets the file's title, author, description, and copyright information. This information is saved in .wav, .aiff, and .mp3 files only. The information is show when you examing the file's "Details" properties in Explorer.

Batch conversion
Batch conversion converts a set of files from one format to another. You can use this feature to compress all your .wav files to .mp3 to save disk space or convert from .mp3 to .wav to prepare for writing to a CD-R disc. To add files to the conversion list, use the Add files button or drag-and-drop files with Windows Explorer. You can remove files by selecting them and choosing the Remove button. The Clear button removes all files from the list. The Rate check box lets you specify a sampling rate for attributes that do not list one, such as 8 or 16-bit files. The rate is not used if the attributes already specify a rate. For some file types, like VOX files, only certain rates are supported, and the closest compatible rate will be used. The destination file types and attributes are the same as given under the Save as command. You must select the file type before selecting attributes since attributes vary from one type to the next. The Destination folder specifies where all the converted files will be stored. If the destination folder is blank, the converted files are stored in the same folder as the original file. The Overwrite existing files option allows you to replace the original files with the converted format. Overwriting only occurs if the destination folder and file type match the original. The Author and Copyright text is stored in file types that support such information as explained under the File | Information command. Press the Begin button to start converting all the files. A status window will appear to show the progress of the conversion and list any errors.

Save
The sound is saved in a file using its original name and type. If memory or disk space is low, the file may not be saved successfully. GoldWave will inform you if this happens. If Save fails, try deleting some unneeded files or close other applications. Make sure that the file is saved successfully before closing GoldWave, otherwise the sound will be lost. Note that audio from video and movie files cannot be saved. You must save those files in an "audio only" format. Cue points are saved only in Wave (.wav) files. If you have added cues to a non-Wave file, you can use File | Save as to convert it to Wave.

Save as
Save as saves a sound using a different filename or file type. To save the sound using a different name, simply type in the new name in the File name box. To save the sound using a different type, select the type from the Save as type list box, then select attributes from the Attributes list box. Since each file type supports different attributes, always select the type before selecting attributes. Java and Web sounds, for example, should be saved using the "Sun (*.au)" type and the "Java/Web" attribute. You can use Save as to compress a sound as well. If you have an MPEG1 Layer 3 encoder installed, for example, you can compress a sound so that it is over 10 times smaller. To do this, select the "MPEG Audio (*.mp3)" type and one of the listed "MPEG" attributes. Use 128kbps (128 kilobits per second) or higher for best quality. If you frequently use a certain file type and attributes, select those settings then choose the Set custom button. The next time you use Save as, you can choose the Custom button to quickly retrieve those settings. Note that just typing in a different extension for the filename does not convert the sound to the type associated with the extension. The extension must be selected from the type list box.

Save selection as
Save selection as saves the selected part of the sound to a file. Use this command to save parts of a large file. The Save as dialog appears where you can specify the new filename, type, and attributes for the file.

Exit
Exit closes all Sound windows and closes GoldWave. Any playback or recording is stopped. You will be asked to save any changed sounds.

File History
A list of several recently used files is appended to the File menu. You can quickly reopen one of these files by selecting it from the menu.

Edit Menu Commands


Edit commands remove, insert, copy, and move sections of sound. For an introduction to the concepts and terms used in this section, refer to the Editing Overview section.

Undo
Undo reverses the most recent change made to a sound. Only one undo is possible across all Sound windows. The undo feature keeps a copy of the original data in a temporary file. This file is created in the undo folder specified by the Options | File command. Since the undo feature copies large amounts of data, you will notice a delay before each modification. It can be disabled using the Options | File command and unchecking the Undo box.

Cut
Use Cut to remove the selection from the sound and put it in the clipboard. The contents of the clipboard can then be superimposed or inserted into a Sound window using Mix or Paste. If you just want to remove the selection and do not need to paste or mix it, you should use the Delete command instead. Note that if only one channel is selected in a stereo sound, then only that channel is removed. Since it is not possible for one channel to be longer than the other, the end of the cut channel is padded with silence (this is also true for Delete). To cut: 1. Move the start and finish markers to the part of the sound you want to cut. 2. Choose Cut from the Edit menu or click on the Cut button.

Copy
The Copy command copies the selection into the clipboard. The selection is not removed from the sound. The contents of the clipboard can then be mixed or inserted into a Sound window using Mix or Paste. To copy: 1. Move the start and finish markers to the part of the sound you want to copy. 2. Choose Copy from the Edit menu or click on the Copy button.

You can copy individual channels of a stereo sound by using the Edit | Channel command to select a single channel.

Copy to
The Copy to command copies the selection to a new file. Use this command to divide a large file into smaller sections. The selection is not removed from the sound. The Save as dialog appears where you can specify the filename, type, and attributes for the file. This is the same as the File | Save selection as command.

To copy the selection to the file "section.wav": 1. 2. 3. 4. 5. Move the start and finish markers to the part of the sound you want to copy. Choose Copy to from the Edit menu. Enter the filename: section Select "Wave (*.wav)" from the type list. Choose Save.

You can save individual channels of a stereo sound by using the Edit | Channel command.

Paste new
Paste new creates a new Sound window containing the sound copied into the clipboard. The new sound will have the attributes and length of the clipboard sound. This command is useful when you need to edit and save part of an existing sound to a new file. To paste part of a sound into a new sound: 1. Move the start marker and finish markers to the part of the sound you want to copy. 2. Choose Copy from the Edit menu or click on the Copy button. 3. Choose Paste new from the Edit menu or click on the PNew button.

Paste and Paste at


After copying a sound into the clipboard, you can use these commands to insert it into another sound. Paste inserts the clipboard at the start marker's position. Paste at inserts the clipboard at the location you specify. The length of the sound is increased so that the clipboard will fit. The clipboard is automatically converted to match the attributes of the sound. To insert the clipboard into the sound: 1. Move the start marker to the place where you want to paste the clipboard sound. 2. Choose Paste from the Edit menu or click on the Paste button. To append the clipboard to the end of the sound: 1. Choose Paste at from the Edit menu. 2. Choose End from the Paste at submenu.

By copying a small selection and pasting it several times, a stutter effect can be achieved.

Mix
Use Mix to blend (combine) the clipboard with the sound. Mixing essentially allows two sounds to be played at the same time, such as vocals and music. You are asked for the volume to apply to the clipboard as it is being mixed. A value of 100 is normal volume. Smaller values make the clipboard sound quieter. Note that before you can use Mix, you need to use the Copy command to copy audio into the clipboard. To mix the clipboard with the sound: 1. 2. 3. 4. Move the start marker to the place where you want to mix the clipboard. Choose Mix from the Edit menu or click on the Mix button. Enter the mix volume for the clipboard. Choose OK.

Replace
Use Replace to replace the selection with the clipboard. The selection is deleted and the clipboard is inserted in its place. If the clipboard is longer or shorter than the selection, the length of the file is adjusted as required.

Delete
Delete removes the selection from the sound. The selection is not copied to the clipboard. You should always use Delete instead of Cut when the selection is not needed. The Delete command is faster because it does not copy the selection to the clipboard. Note that if only one channel is selected in a stereo sound, then only that channel is removed. Since it is not possible for one channel to be longer than the other, the end of the deleted channel is padded with silence (this is also true for Cut). To delete: 1. Move the start and finish markers to the part of the sound you want to delete. 2. Choose Delete from the Edit menu or click on the Delete button.

Trim
Trim removes everything outside the selection. The selection is not affected. Use this command to keep a section of sound and discard everything else. This command is frequency used after recording. Note that if only one channel of a stereo sound is trimmed, the end of that channel will be padded with silence. As an alternative, you can use the Copy to command to save the selection to a separate file. To trim: 1. Move the start and finish markers to the part of the sound you want to keep. 2. Choose Trim from the Edit menu or click on the Trim button.

Insert silence
This command inserts some blank space in the sound at the start marker's position. You are asked how long (in seconds) the silence should be. This command can be used to increase recording time or to insert a delay.

Select view
Use Select view to select all of the sound currently shown in the Sound window's graph. The start and finish markers are moved to the far left and far right of the view. This command appears on the tool bar as the View button.

Select all
Use Select all to select the entire sound. The start and finish markers are moved to the beginning and end of the sound.

Channel
The Channel submenu sets which channel of a stereo sound will be used or modified by editing or effects. You can use this feature to copy a single channel from a stereo sound or apply an effect to only one channel. The currently selected channel is shown in the status bar. When recording or using effects such as the Expression evaluator, Resample, Playback rate, Pan, and Exchange channels, the channel setting has no effect and both the left and right channels are modified.

Marker
This submenu lists commands for changing the positions of the start and finish markers.

Set
Sets the start and finish marker to an exact time or sample position. To specify a time, choose the Time option and enter the time in hours, minutes, seconds, and thousandths of a second. For example, you could enter 1:04:27.873. To specify a sample position, choose the Sample option and enter the position. If you want the length of the selection to be aligned to a CD sector or 1 kilobyte, select the appropriate option. When the OK button is pressed, the finish marker will be adjusted to align the selection length.

Recall positions
Moves the start and finish markers to previously stored positions. These positions are set using the Store positions command.

Store positions
Saves the current positions of the start and finish markers. Use the Recall positions command to move the markers back to these positions.

Drop start/finish
When playing a file, you can drop the start or finish marker at the current playback position. You can use the bracket keys, [ and ], to perform the same command. Note that the start marker cannot be dropped after the finish marker.

Snap to zero-crossing
When editing, it is necessary that the waveform not change suddenly from one sample to the next, otherwise a click will occur. This can happen when deleting the selection. The amplitude of the waveform at the start marker may be completely different from the amplitude at finish marker. After deleting the selection, these two different amplitude will be right next to each other, causing a click. The Snap to zero-crossing feature helps to minimize the problem by making sure that the markers are always near zero amplitude samples. When you drag and release a marker, it is automatically moved to a position where the amplitude approaches zero. This means that when you delete the selection, the amplitudes at both the start and finish markers will be more closely matched (near zero). Since stereo sounds can have very different left and right channels, it is not always possible to find an ideal zero-crossing position. However, you can use the Edit | Channel submenu to limit the snap feature to a single channel.

Deflash
Deflash copies a "flash opened" file to temporary storage. Usually a file is deflashed automatically. If you are trying to play a compressed file on a slow system, convert a file to another type, or directly editing a waveform with the mouse, use this command to copy the file for processing. The flash feature can be configured using the Options | File command. See the Storage Overview section for more information.

Effects Menu Commands


With Effects commands, you can dramatically enhance and change sounds. These commands are similar to font menu commands in word processors. For example, using font commands, you can change the size of the letters. In GoldWave, using the Volume effect changes the "size" of a sound. Changing the colour of a font would be similar to changing the pitch of a sound. Note that even though the word "volume" is used throughout this section for readability, "amplitude" would be more precise. For an introduction to some of the terms used in this section, refer to the Editing Overview section and Appendix A.

Special Controls for Effects


Many effects have similar controls such as presets and shape boxes. These are explained below.

Presets
Presets store parameters and shapes (described below) in the gwpreset.ini file so they can be recalled again the next time the effect is used. Controls for presets consist of a drop down list box, a [+] add button, a [-] remove button, and sometimes a [/] clear button, as shown in Figure 8. To add a new preset: 1. Enter in all the new parameters and/or draw the new shape. 2. Type in a new name for the preset in the drop down list. This name cannot be the same as one currently in the list. 3. Choose the [+] button. To delete a preset: 1. Select the preset from the drop down list. 2. Choose the [-] button. To change a preset: 1. Delete the preset, as above. When you delete a preset, the current parameters and name remain on the screen so they can be changed. 2. Change the parameters and/or name. 3. Add the preset, as above.

Figure 8: Special Controls for Effects

Shape Controls
Several effects in GoldWave use Shape Controls to set graphical parameters or dynamically alter the effect across the selection. Shape Controls usually consists of a graph window and presets. The graph window initially contains a single line with two endpoints (shown as large dots). By clicking the left mouse button anywhere inside this window, you can add new points to bend the line into a variety of zigzag shapes. To move a point, click on it and drag it to a new location. To remove a point, click the right mouse button over the point. The clear button removes all the points and reset the end points. Note that endpoints cannot be removed.

Doppler
A Doppler effect is defined as a change in frequency of a wave caused by motion. It is often heard during auto racing when a fast car passes in front of you. The pitch of the engine appears to drop as the car speeds away. In GoldWave the Doppler command dynamically alters or bends the pitch of the selection. Shape Controls are presented where the pitch can be varied over the selection from 0.5 to 1.5 times normal. You can use Effects | Volume | Shape to dynamically alter the volume as well. The "Power loss" preset gives you a good idea of what it sounds like when the batteries start to fail in a portable tape player. Other presets can change your voice to a smurf or a giant.

Dynamics
Dynamics alters the amplitude mapping of the selection. It can limit, compress, or expand a range of amplitudes. The amplitude mapping is set using Shape Controls, where x-axis and yaxis both have a range of -1 to 1. When the line stretches diagonally from the lower left corner to the upper right corner, the input amplitude (x) and output amplitude (y) are the same for every point on the line. By changing the line, the output will differ from the input.

Figure 9: Dynamics Figure 9 shows an example of amplitude mapping for clipping distortion. Point P1 has an input value of -0.4 and an output value of -0.4. Therefore no change occurs to the amplitude. Point P2 on the other hand, has an input value of 0.8 and an output value of 0.5. In this example, all input amplitudes in the range of -0.5 to 0.5 remain unchanged. Any values outside this range will be limited to 0.5, so that the final sound will have no amplitude magnitudes greater than 0.5. Essentially, any values that are too high are "clipped" to fit within the range.

In practical terms, dynamics can increase the volume of quiet sections of a sound without greatly increasing the loud sections as well. It can introduce mild or heavy distortion effects (such as the "Blare" or "Level noise" preset).

Echo
Echo produces an echo or reverb effect in the selection. The echo delay, volume, and reverb parameters can be entered after choosing this command. The delay determines how long it takes for the echo to bounce back. Try values less than 0.1 for a large room, 0.3 for a baseball stadium, above 0.5 for a canyon echo. The volume controls how loud the echo will be. Values less than 50 give good results. Reverb makes the echo sound deeper and richer. If you check the Reverb box, the echo will be regenerated at intervals specified by the delay. This means that if the delay was 0.1 seconds, the echo at 0.1s is regenerated at 0.2s, and this new echo is regenerated at 0.3s, and so on. The volume is applied to each regeneration. If the volume was 50%, the first echo volume is one half the original, the second echo volume is one quarter, and so on. To make the echo sound correct, the effect extends slightly outside the end of the selection. This may increase the length of the sound or alter sound outside the selection. To add an echo: 1. 2. 3. 4. 5. 6. Move the start and finish markers to the part of the sound you want to add an echo. Choose Echo from the Effects menu. Enter the delay time. Enter the volume. Check Reverb if preferred. Choose OK.

Expand/Compress
The Expand/Compress effect is a general purpose dynamics processor that includes compressors, limiters, expanders and gates. Compressors and limiters are used to decrease or limit the dynamic range of audio. They reduce the volume of loud sounds while leaving the rest of the sound unchanged. Expanders and gates are used to increase the dynamic range of audio. They reduce or eliminate quiet sections, which can help to reduce background noise. Compressors always work on loud sections and expanders always work on quiet sections. Normally, both compressors and expanders only reduce the volumes. However, GoldWave also allows you to boost the volume. The Ratio specifies the compression or expansion ratio. For compression, this value should be less than 100%, typically between 25% and 75%. It essentially defines the scale factor of the loud volumes. For a limiter, use a value less than 10%. For expansion, the value should be less than 100% as well. It defines the scale factor of the quiet volumes. To boost the quiet volumes, use a value greater than 100%. For a gate, use a value of 0%. For those with a technical background, note that this is the opposite of how a standard expander works, but it makes the ratio more consistent between compression and expansion. This way, a value less than 100% always decreases volumes and a value greater than 100% always increases them.

The Threshold specifies the envelope level to activate the expander or compressor. Compressors change the volume level of all sounds above that level. Expanders change the volume level of all sounds below that level. Depending on the Smoothness setting, the threshold may have to be set much lower than expected. The Smoothness specifies how quickly the compressor/expander changes from one volume level to the next and how quickly it activates. Using 0% means that volumes will change instantly, which can cause a rough distortion in sections of audio that border on the threshold level. A value of 100% means that volumes will change gradually over 100ms. With a high smoothness setting, the threshold will have to be reduced. The higher setting makes the envelope detector respond more slowly to changes in the sound, resulting in a lower envelope range. Use the Expander and Compressor options to specify what processing is required. Compressor Example You have recorded some music that has a few loud moments, but you want to raise the overall volume without distorting the loud parts. With the Compressor option selected, use 25% as the ratio, 0.200 as the threshold and a smoothness of 50%. To compress loud volumes even more, lower the ratio or lower the threshold. After compression, use the Volume | Maximize command to stretch the volume to the full dymanic range. Expander Example You have recorded someone talking and notice background noise during the quiet parts. To reduce the noise, use the Expander option, 5% as the ratio, 0.050 as the threshold and a smoothness of 50%.

Filter
Filters are used to remove a range of frequencies from a sound and can produce a variety of effects. A submenu is displayed listing several filter related commands.

Noise gate
Noise gates remove background hiss or noise from quiet parts of the selection. You can use this after recording to clean up some of the noise created by the audio device when it converted the sound to digital data. This works well for voice recordings where there are frequent "silences" between words. Noise gates do not remove background hiss from louder parts of the selection, making them unsuitable for music. The Attack time is the amount of time (in milliseconds) that it takes for the noise gate to fully close. When the gate is closed, no sound can pass and this leaves only silence. Start with a value of about 200 milliseconds or less. The Release time is the amount of time (in milliseconds) that it takes for the noise gate to fully open. A value of 50 or less usually gives good results. The Threshold is the amplitude level at which the gate will start to open and let sound pass. If you specify a value of 0.05, for example, all samples with levels from 0.05 to 1.0 will be allowed to pass. Samples with levels 0 to 0.05 are blocked. If you still notice a hiss in quiet sections, increase this value and decrease the attack time.

The Anticipation settings allows the noise gate to predict when the gate should open or close. This significantly reduces the problem associated with analog noise gates where the beginning of the sound is chopped off during the time it takes to release the gate. A value of 10 makes the noise gate look 10 ms head. In general, it is best to set this value equal to the Release time.

Noise reduction
Noise reduction uses frequency analysis techniques to remove unwanted noise from a sound, such as a background hiss, a power hum, or random interference. It cannot be used to separate or remove complex sounds, such as removing vocals from music or a cough from a speech. The interface (Figure 10) includes a frequency analysis window, with a shape line, and several other controls. The Coordinates group show the x and y coordinates when you click-and-drag a shape point. The x coordinate is the frequency in Hertz and the y coordinate is the magnitude in decibels.

Figure 10: Noise Reduction The time of the frequency analysis is given as T, in seconds. If you adjust the scroll bar located below the analysis window, the analysis time can be changed to show the frequency analysis of a different part of the sound. The width of the analysis depends on the FFT size setting, explained below. Noise is removed using a reduction envelope. The shape of the envelope should closely match the shape of the noise you want to remove. The frequency analysis graph can help determine that shape. Adjust the analysis time so that it coincides with a time in the sound where only the noise is heard (play the file to find such a place and time). Once you have isolated the noise in the analysis graph, you then create the envelope. The envelope can be created in four different ways, depending on the Reduction envelope setting: Use shape Lets you manually create an envelope shape or select a preset shape. See Shape Controls for information about creating shapes. By creating a horizontal line at about 75 dB, you can remove a hiss from a sound.

Use current spectrum Creates an envelope based on the shape of the graph shown in the frequency analysis window. This is particularly useful for removing a complex buzz or hum. Use average Applies an averaging envelope throughout noise reduction processing. The envelope is continuously updated, based on the frequency analysis of the sound. Use this setting if the noise changes frequently throughout the sound. Use clipboard Creates an envelope based on an analysis of the waveform in the clipboard. This is the most flexible option and usually gives the best results. Before you can use this option, you must use Edit | Copy to copy a piece of noise into the clipboard. For best results, the piece should contain only the noise you want to remove from the rest of the file. The noise can even be copied from a different file. After you copy the noise, remember to change the selection to the part of the file you want to apply the noise reduction. Settings The FFT size determines the detail of the frequency analysis and the noise reduction envelope. Usually values between 9 to 11 give the best results. The Overlap value specifies the percentage of the FFT size to overlap from one calculation to the next. A value of 75 is best. The Scale value lets you alter the reduction envelope scale. A value of 100 uses the envelope as it is. A value of 200 doubles the envelope, which doubles the amount audio removed from the sound. A value of 50 halves the envelope, which halves the amount removed. Normally it should be set to 100.

Low/Highpass
Lowpass filters block high pitched frequencies (treble), but allow low pitched frequencies (bass) to pass. They can be used to reduce high end hiss noise or remove unwanted sounds above the given cutoff frequency. If you were to apply a lowpass filter with a cutoff frequency of 1000 Hz on speech, it would make it sound mumbled and deep. Lowpass filters can also be used to eliminate aliasing noise when used before downsampling. Highpass filters block low pitch frequencies, but allow high pitched frequencies to pass. They can remove deep rumbling hum or remove unwanted sounds below the given cutoff frequency. If you were to apply a highpass filter with a cutoff frequency of 1000 Hz on speech, it would make it sound thin and hollow. Cutoff frequency The Initial box specifies the constant cutoff frequency for static filtering. If the Dynamic option is selected (see below), then a final cutoff frequency can be given in the Final box. Filter options Select Lowpass if you want to keep only the frequencies below the cutoff frequency. Select Highpass if you want to keep only the frequencies above the cutoff frequency.

If you want the cutoff frequency to remain constant throughout the selection during processing, select the Static option. If you want the cutoff frequency to change from the initial value to the final value, select the Dynamic option. Note that dynamic filtering will take more processing time. The Steepness value specifies how sharply the filter cuts off frequencies outside the cutoff frequency. A higher steepness makes the filter sharper, but it also increases processing time. In technical terms, the steepness specifies the number of second order cascade filters used. Examples To make speech gradually become more hollow and thin: 1. 2. 3. 4. 5. Enter 60 in the Initial box. Choose Dynamic. Enter 1000 in the Final box. Choose Highpass. Choose OK.

Filtering before downsampling from 44100 Hz to 22050 Hz: 1. 2. 3. 4. 5. Enter 11025 in the Initial box. Choose Lowpass. Choose Static. Enter 20 in the Steepness box. Choose OK.

Bandpass/stop
Bandpass filters block all frequencies outside the specified range, keeping only frequencies within the range. Bandstop filters block all frequencies within the specified range, keeping all other frequencies outside the range. Frequency range The From and To boxes specify the frequency range of the filter. If the Dynamic option is selected, then a final frequency range can be given in the other From and To boxes. Filter options Select Bandpass if you want to keep only the frequencies within the range. Select Bandstop if you want to keep only the frequencies outside the range. The remaining options are explained above under Low/Highpass filters.

Equalizer
Equalizers are commonly found on stereo systems. They boost or reduces certain ranges of frequencies. Simple equalizers control only treble and bass. GoldWave's equalizer (Figure 11) controls 7 bands.

Figure 11: Equalizer Center frequencies for each of the 7-bands are given at the top of each scroll bar. Adjust the scroll bars to boost or reduce a band by +12 dB to -24 dB. To change bass, adjust the two or three left-most bands. To change treble, adjust the two or three right-most bands. Several presets are included to demonstrate bass and treble changes. More detailed equalization is possible using the Parametric EQ, described below.

Parametric EQ
A parametric equalizer (Figure 12) is a flexible tool for reducing or enhancing ranges of frequencies. GoldWave presents an easy to use interface where all the parameters for up to 30 band can be configured quickly. The Presets contain some commonly used settings.

Figure 12: Parametric Equalizer Graph window The graph shows frequency on the x-axis in Hertz and the gain on the y-axis in decibels. Each enabled band is displayed in the graph as a diamond shaped box located at its center frequency and gain. The width of the box shows the bandwidth. The currently selected band is shown in blue and its exact settings are given in edit box controls. A short time frequency analysis graph is drawn with the left channel in green and the right channel in red. The time of the analysis can be changed using the scroll bar located at the bottom of the graph. The analysis can be used to determine what frequencies to attenuate or intensify. Controls A band is configured by selecting its number from the Select band box and adjusting the scroll bars. A quicker way is to drag-and-drop the band to a new location on the graph. Note that because of the logarithmic frequency scale, the width of a diamond changes as you move it left or right. The bandwidth, however, remains constant. Any unneeded bands can be disabled by unchecking the Enabled box. Disabling unused bands improves processing speed. The "Notch" preset is effective for removing a specific tone from a sound, such as a 60 Hz hum. The "Bass boost" and "Treble boost" presets work the same way as the bass and treble controls on a stereo system. Just adjust the gain up or down to control them.

Pop/Click
A pop/click filter is a specially designed filter that searches for abrupt changes in the sound and eliminates them. Such a filter is often used to remove pops and clicks caused by dust and scratches when recording old vinyl records.

The Tolerance defines how abrupt a change can be before it is considered a click. It is best to start with a value near 1000%. Using a lower value will detect more clicks, but may eliminate natural clicks such as drum sticks tapping together or a conductor tapping the baton. Values less than 500% should be used on short selections only. When a click is detected, the filter attempts to reconstruct the damaged waveform based on the surrounding waveform shape making the repair almost imperceptible. However with excessive pops and clicks or at low tolerance levels, reconstructed waveforms may overlap and sound distorted. The tolerance setting should be kept as high as possible. Using a very low setting may introduce more distortion than existed in the original. The filter requires a minimum selection of 4000 samples (about one tenth of a second at CD quality) to establish a base line. Using the filter on a shorter selection has no effect.

User defined
The User Defined Filter dialog allows you to specify coefficients to use for filtering. Up to 15 coefficients can be given. Almost any kind of linear filter can be created with this command because it exploits the general digital filter equation:

In GoldWave, this becomes: b(0)y(n)+b(1)y(n-1)+ ... +b(14)y(n-14) = a(0)x(n)+a(1)x(n-1)+ ... +a(14)x(n-14) The number of coefficients entered for a and b must be the same. For FIR filters, you would usually enter a one followed by a number of zeros for b. You can use Ctrl+C and Ctrl+V to copy and paste coefficients in the edit boxes. The Clear button quickly removes all coefficients. Some predefined filters are included in the Coefficient Sets. The number following a lowpass filter preset indicates what percentage of frequencies are kept. Lowpass 25, for example, keeps the lower 25% of frequencies. The number following a highpass filter preset indicates the percentage discarded. Highpass 10, for example, discards the lower 10% of frequencies. The actual frequencies kept or discarded depends on the sampling rate of the sound. Lowpass 25 on a 22050 Hz sound will remove frequencies from about 2700 Hz to 11025 Hz. To fully use this command requires detailed knowledge of digital filter theory, which is beyond the scope of this manual. A brief introduction is provided in Appendix C.

Flange
A flanging effect is similar to an echo effect in that the original sound can be mixed with a delayed copy of itself. Unlike an echo, where the delay is constant, flanging varies the delay over a specified range or depth. The speed, or frequency, at which the delay varies can be controlled as well. In GoldWave, the Flange effect presents a dialog where you can set the depth, frequency, and fixed delay parameters and define how the sound should be mixed. Several

presets are included to demonstrate the kinds of unusual audio effects that are possible with flanging. The Input volume specifies the volume of the original sound to mix with the final sound. A value of 0 means the original sound will not be mixed with the final sound. If this value is set to 100 and all other volumes are 0, no change will be made to the sound. A value of -100 simply inverts the input, which is equivalent to subtracting the original instead of adding it to the final sound. The Mix volume specifies the volume of the flanged (delayed) sound to mix with the final sound. Usually, this value should be in the range of 50 to 100, or -50 to -100 for an inverted mix. Feedback specifies the level of feedback to mix with the final sound. This makes the effect sound more pronounced. Set this value to 0 if you do not want any feedback. In general the feedback should be set to between -75 to 75. Depth specifies the maximum variable delay in milliseconds. A value of 40 will allow the delay to vary from 0 to 40 milliseconds. Frequency specifies how fast to vary the delay. A value of 2 will vary the delay over its depth twice a second. For a value of 0.2, the full delay depth is reached every five seconds. The Fixed delay is added to the depth to change the minimum delay. If the depth is 40 and the fixed delay is 10, the delay will vary from 10 to 50 milliseconds.

Interpolate
Interpolate (Figure 13) uses linear interpolation to smooth out samples between the start and finish markers. Use this command on a tiny selection to remove a pop or click.

Figure 13: Interpolate

Invert
Invert reflects the selection about the time axis. The selection is essentially turned upside-down. This produces no noticeable effect in mono sounds and has a slight effect in stereo sounds. Inverting a single channel of a stereo sound produces an "in" or "out" effect. Inverting can be used before mixing so that the two sounds are subtracted instead of added.

Mechanize
Mechanize adds a robotic or mechanical characteristic to sounds. The percentage of quality can be entered after selecting this command. Low values produce an untuned radio effect. Higher values give a rough distorted effect.

To mechanize part of a sound: 1. Move the start and finish markers to the part of the sound you want to mechanize. 2. Choose Mechanize from the Effects menu. 3. Enter the quality percentage, then choose OK.

Offset
Offset adjusts or removes a dc offset in the selection by shifting it up or down (Figure 14) so that the wave is centered on the horizontal axis.

Figure 14: Offset

When this command is selected, it first scans the selection for any existing offset. An offset to cancel the existing one is then displayed in a dialog where it can be changed. A positive value shifts the graph up and a negative value shifts it down. If a value of 0 is displayed, the sound does not have an offset. An offset in a waveform should be removed to minimize pops/clicks during editing. Offsets may adversely affect other effects. To adjust the offset of part of a sound: 1. Move the start and finish markers to the part of the sound you want to adjust. 2. Choose Offset from the Effects menu. 3. Enter the offset, then choose OK.

When removing an offset from a stereo sound, use the Edit | Channel submenu to change each channel separately. Stereo sounds often have different offsets on the left and right channels. You should check the offset from time to time after processing effects. Otherwise, the offset may increase with each effect, resulting in distortion.

Pitch
Pitch changes the pitch (frequency) of the selection. This is useful for converting instrument samples from one note to another. The new pitch is specified using a scale factor or using semitone and fine tune values.

Scale This option scales the pitch by the value you specify. If you set the scale to 0.5, that will be equivalent to a downward shift by one octave. A value of 2.0 is the same as an upward shift of one octave and would make a voice sound like a chipmunk. A value of 0.75 would make a woman's voice sound like a man's. Semitone This option changes the pitch by semitones (notes on a piano). If your sound is a note at middle C and the semitone value is 2, the note will be changed to D. A value of -1 changes the note to B. A value of 12 make the note one octave above middle C. The Fine tune value lets you make a slight pitch adjustment in hundredths of a semitone. For example, a value of 50 changes a note from C to halfway between C and C#. Preserve length If this option is checked, a complex algorithm will be used to keep the length of the original note the same as the new note. In other words, the tempo will not be changed. In terms of a voice recording, this changes the pitch of the voice without changing the speed at which the words are spoken. This option requires a substantial amount of processing time and will affect the quality of the sound. The FFT size determines how much of the sound to process at one time, based on a power of 2. A value of 10, for example means that 1024 (210) samples will be processed at a time. Values from 9 to 11 give good results. The Overlap determines what percentage of the processed samples will be recalculated. It should be at least 88. Values of 90 and 95 will give better results, but require more processing time.

Reverse
This command reverses the selection so that it plays backward. Now you have an easy way to listen to all those "satanic verses" or reverse speech messages. You can play a sound backwards by using the rewind button on the Device Controls window as well.

Silence
The Silence command erases the selection. The sound in the selection is replaced with silence.

Stereo
The Stereo submenu contains commands that apply to stereo files, such as swapping channels and left/right panning.

Exchange channels
This command exchanges the left and right channels of a stereo sound (i.e. the right channel becomes the left channel and the left channel becomes the right channel).

Pan
Pan presents the Shape Controls where left and right panning can be controlled. The graph is divided into green and red regions, representing the left and right channels respectively. The line, initially located between the regions, represents the center for panning. By bending and/or moving the line, you can dynamically alter the selection's left/right balance or pan to and from each channel. Figures 15 to 17 show several examples of panning shapes.

Figure 15: Pan from left Figure 16: Pan from right to left and Figure 17: Pan from left center to to right back to right right center

Remove vocals
This effect removes vocals from certain stereo recording by subtracting the left and right channels. This works only when vocals are located in the exact center of the stereo image. Note that any instruments located in the center will be removed as well. After processing, the stereo image is lost and the final output will sound monaural.

Time warp
Time warp (Figure 18) changes the playback speed or alters the tempo of the selection. This effect has many uses: it can stretch or compress a sound to fit in a certain time, it can slow down instrumental music for easy transcription, or it can change the tempo of one musical passage to match rhythm and beats of another. Three different techniques are provided, each with certain advantages and disadvantages. All three let you specify the change either by a speed factor or by a new length. The Speed factor lets you specify a relative change. A value of 0.5 make the selection play twice as slow. A value of 2.0 make the selection play twice as fast. The Time options lets you specify a new length for the selection. This is useful if you need to make a sound fit a certain time, such as squeezing a 35 second commercial into a 30 second spot.

Figure 18: Time Warp

Speed Speed changes the sampling rate of the entire sound so that it plays back at a different speed, similar to spinning a vinyl record faster or slower. It works the same way as the speed scroll bar in the Device Controls window, but in this case, the sound itself is changed. This technique is very fast and produces excellent quality, however, the pitch of the sound is changed as well. In other words, if you were to speed up a voice, the pitch becomes higher, making the voice sound like a chipmunk. Similarity Similarity uses correlation to add and overlap small, similar sections of the sound. This technique preserves the pitch. It generally produces high quality voice and fair quality music when using small speed or time changes. A fair amount of time may be require for processing, depending on the Search range value. For voice, the Window size should be set between 20 and 30 and the Search range set to between 5 and 10. For music, a larger Window size and Search range gives better results, such as 50 and 25. FFT FFT uses Fourier transforms and interpolates or decimates the frequency analysis to change the length. This technique preserves the pitch, but can introduce some artifacts into the sound. Best quality is obtained by using the Oscillator synthesis option, but that requires significant processing time. The FFT size should be set from 9 to 11 and the Overlap should be at least 75, but can be set to 88, 90, or 95 for better quality. The FFT size and Overlap settings are explained under the Noise reduction filter section. If you changed the Device Controls playback speed, remember to set it back to 1.00 so that the device plays at the correct speed.

Volume
The Volume submenu contains several volume related commands. Volumes are usually specified by a percentage of the sound's original amplitude. A value of 200 doubles the volume and a value of 50 halves the volume. Unlike the volume scroll bar in the Device Controls window, which changes the audio device output volume, these commands alter the sound's data to change the volume. To convert gain in dB to a percentage, use the formula:

Change
This command modifies the selection so that it sounds louder or quieter. You need to enter the new relative volume. Values less than 100 make the selection quieter. Values greater than 100 make it louder. A value of 100 is normal volume and has no effect. If you are trying to make the volumes of several different songs sound the same, use Effects | Volume | Maximize to obtain the rms value for each song. You can then use this effect

to scale the volume by the appropriate amount. If the required rms value is 0.25 and it currently is 0.20, the change would be 0.25 / 0.20 = 1.25 or 125%. To double the volume of part of a sound: 1. Move the start and finish markers to the part of the sound you want to change. 2. Choose Change from the Effects | Volume submenu. 3. Enter 200 for the volume percentage, then choose OK.

Fade in
Fade in gradually increases the volume throughout the selection. You need to specify the initial volume percentage. A value of 25 starts with one quarter volume and fades in to full volume. A value of 0 starts at silence and fades in to full volume. To fade in part of a sound from silence: 1. Move the start and finish markers to the part of the sound you want to fade in. 2. Choose Fade in from the Effects | Volume submenu. 3. Enter 0 for the initial volume, then choose OK.

Fade out
Fade out gradually decreases the volume throughout the selection. You need to specify the percentage of fade. The fade percentage is the amount that the volume should decrease. A value of 100 fades to complete silence. A value of 50 fades to half the original volume. To completely fade out part of a sound: 1. Move the start and finish markers to the part of the sound you want to fade out. 2. Choose Fade out from the Effects | Volume submenu. 3. Enter 100 for the fade percentage, then choose OK.

Maximize (Normalize)
Maximize searches the selection for the current maximum volume level. It then displays the level, the position of the level within the file, and the RMS (root-mean-square) level. You can then specify a new maximum volume level. The volume of the entire selection is changed so that the maximum will match that value. This is often referred to as "normalizing". The RMS value indicates of the overall volume of the selection. If you want two separate sounds to have a similar volume, you can adjust the RMS value of one sound so that it matches that of the other. To choose the best RMS value to use, scan each sound and either calculate the overal average or just choose the minimum. Choosing the minimum RMS value ensures that no clipping distortion will result from the change. To maximize to full volume a part of a sound: 1. Move the start and finish markers to the part of the sound you want to normalize. 2. Choose Maximize from the Effects | Volume pop-up menu. 3. Enter 1.0 for the new maximum, then choose OK.

Shape
Shape presents the Shape Controls where the volume envelope of the selection can be defined. The shape line is initially horizontal at 100, representing normal volume. By bending or moving the line, you can dynamically change the volume over the selection. Adding a point below 100 decreases the volume. Adding a point above 100, in the red section, increases the volume. Note that increasing the volume may cause clipping distortion. Several preset shapes are included. The "Exp fade out" and "Exp fade in" presets demonstrate customized fading.

Playback rate
This command changes the playback rate of the entire sound. The sound will play faster (or slower) and its pitch will be higher (or lower). Essentially, this just changes the first number in the status bar. Values of 11025, 22050, and 44100 are recommended. To change the playback rate of the entire sound: 1. Choose Playback rate from the Effects menu. 2. Enter the new rate, then choose OK. The playback rate of the audio device can be controlled using the speed fader in the Device Controls window.

Resample
Resample changes the sampling rate of the entire sound. Unlike Playback rate, this command re-calculates and interpolates all the data so that the pitch and playback time are not affected. This command is useful for converting any sampling rate to the standard CD rate of 44100 Hz or the standard telephony rate of 8000 Hz. You are prompted to enter a new rate. To change the sampling rate of the entire sound: 1. Choose Resample from the Effects menu. 2. Enter the new rate, then choose OK. If you have a sound recorded at 44100 Hz and do not need CD quality, you can save large amounts of disk space by resampling the sound to 22050Hz or 11025Hz. This reduces the size by 2:1 or 4:1. Before down-sampling (converting 22050 Hz to 11025 Hz, for example), the data should be lowpass filtered to prevent aliasing. See Effects | Filter.

View Menu Commands


This section assumes that you are familiar with the terms introduced in the Interface Overview and Editing Overview sections. View commands allow you to see a more detailed graph of part of the sound. They are similar to zoom commands in the Windows Paint accessory. When you zoom in (or magnify) the sound, you see a smaller section, but with greater detail. When zoomed out, you see the entire sound, but with less detail. The Overview box near the bottom of each Sound window gives you some information about what section of the sound is currently shown in the view (see Figure 1).

When zoomed in to a part of the sound, a scroll bar will appear at the bottom of the Sound window so you can move to different parts of the sound while still keeping the same level of magnification. The current level of magnification is displayed in the Main window's status bar next to the word "Zoom". Most view commands use the start marker's position as the starting location for magnification, so you should move the start marker to the position of interest first. Note that if a file is flash opened, only part of the sound is initially display in the view. Otherwise, the entire sound is displayed.

All
The entire sound is graphed in the view. In other words, it zooms all the way out so that the entire sound is visible. You can move the start and finish markers to select any part of the sound.

Other
This magnifies the graph to any level you specify. The level is given as 1:X, where X is the number you enter in the box. A value of 10 gives a 1:10 level as described below. A value of 0.10 is equivalent to a detailed 10:1 zoom level. If the given level is not possible, the closest valid level is used.

Previous zoom
This returns the view to the previous zoom level. Use this to switch back and forward between two different zoom levels.

Selection
The selection is magnified, increasing the detail of the graph (Figure 19). You can zoom in many times by changing the selection and magnifying it again until only a single sample is shown in the view.

Figure 19: View Selection

User
A User button is provided in the tool bar so that you can quickly zoom to your favourite level. The sound is magnified to the level of detail specified under the Options | Window dialog. The level can be set to any value you find convenient.

Zoom in
Magnifies the sound by a factor of 1.33x. This gives 33% more detail, but shows 33% less sound. The middle of the view is used as the zoom focus. This command complements Zoom out.

Zoom out
Reduces magnification by a factor of 1.33x. This gives 33% less detail, but shows 33% mode sound. The middle of the view is used as the zoom focus. This command complements Zoom in.

Zoom 10:1 and 5:1


When the number to the left of the colon is greater than one, a very small section of sound is magnified at a high level of detail. At these levels, individual samples are easily visible and direct waveform editing with the mouse is possible.

Zoom 1:1
At a level of 1:1, each audio sample is represented as a single pixel on the screen. This reveals a true representation of the shape of the sound.

Zoom 1:10, 1:100, 1:1000


When the number to right of the colon is greater than one, a larger section of sound is displayed, but with less detail. Levels beyond 1:10 show only an approximation of the shape of the sound with minimum detail.

Vertical zoom all


Vertically zooms all the way out so that the entire vertical amplitude range of the sound is shown.

Vertical zoom in
Magnifies the graph vertically to show 1.33x as much amplitude detail. Zooming is focussed on the horizontal center of the view.

Vertical zoom out


Reduces vertical magnification to show 1.33x less amplitude detail. This show a larger range of the amplitude. Zooming is focussed on the horizontal center of the view.

Start and Finish


These commands scroll the view to either the start or finish marker's position. The view will be centered over the marker's position provided its position and the level of magnification permit it to be centered. These commands are especially useful when you need to move a marker to a precise position. For example, you can zoom in 1:1 and move the start marker to an exact position and then use View | Finish to set the finish marker's position.

Tools Menu Commands


Cue points
Cue points mark and describe specific positions within sounds. They have numerous uses. When recording speech, for example, you can use them to hold information about the speaker or a translation of what the speaker said. For music, you can store lyrics for each verse. If you design instrument samples, cue points can hold looping points. Some multimedia applications use them to play or loop specific sections of a sound. When transferring albums to CDs, cue points can mark track division points, which can be used later to divide a large file into individual songs or tracks. Cue points can be set at the start or finish marker's position. You can also move the start or finish marker back to a cue point. You can drag-and-drop a cue point using the left mouse button. Clicking the right mouse button on a cue point displays a context menu. Currently, cue points are fixed and do not change position when a sound is modified. This should be considered when certain commands, such as delete, are used. Any cues inside the deleted selection will not be deleted and the cues outside the selection will not be adjusted to account for the new positions. Cue points are saved only in Wave files. The Split file feature divides a large file into smaller files by using the cue points as split points. This can be used to divide a recording of an album into individual songs before writing to a CDR disc as audio tracks. The smaller files are stored in the specified Folder. The Base name is used together with a 3 digit number to create the name of each file. If you entered "Track", for example, the files would be named Track000, Track001, Track002, etc. The Overwrite existing files option automatically replaces files with the same name that already exist in the folder. The Use CD format and alignment ensures that each file is stored in CD compatible format and that the length of the audio is exactly aligned to a CD sector boundry. The Mark silence button automatically searches for and sets cue points at regions of silence. This is very useful for marking the silence between songs so that a file can be split into individual songs. When regions of silence are found that match the given settings, cue points are set in the middle of each region. The Silence threshold setting controls what level should be considered silence. Typically this should be 0.05 or 0.1 for most recordings, depending on the background noise. The Minimum length setting controls how long the silent region must be before it is marked. To find separations between songs, values from 1 to 1.5 seconds are best. Applying pop/click and noise reduction filters are recommended before using this feature. To set a cue point at the start marker's position: 1. 2. 3. 4. 5. 6. Move the start marker to the position where you want to create a cue point. Choose Cue points from the Tools menu. Choose the Start marker radio button. Enter a name and description for the cue point. Choose the Add button. Choose OK.

To delete a cue point: 1. 2. 3. 4. Choose Cue points from the Tools menu. Select the cue point from the list. Choose the Delete button. Choose OK.

To change a cue point: 1. 2. 3. 4. 5. 6. 7. Move the start marker to the position where you want to move the cue point. Choose Cue points from the Tools menu. Select the cue point from the list. Choose the Start marker radio button. Change the name and description, if necessary. Choose the Revise button. Choose OK. If you do not want to change the cue point's position, you should move the start marker to the cue point first by using the Move marker button. To move the start marker to a cue point: 1. 2. 3. 4. 5. Choose Cue points from the Tools menu. Select the cue point from the list. Choose the Start marker radio button. Choose the Move marker button. Choose OK.

Expression evaluator
The Expression Evaluator is a comprehensive tool for manipulating and generating audio data. For a detailed explanation, see Appendix C.

CD player
If you installed the CD player bundled with Windows, you can use this command to start it. Note that the command will not be enabled if the CD player is not installed. For information about using or installing the CD player, refer to the Windows help.

Volume controls
This command starts the Volume Control accessory, if it is installed on your system. If this accessory is not installed, the command is disabled. The accessory initially shows volumes for output devices. To show recording volumes, use Options | Properties on the menu, choose the Recording radio button, and make sure that the sources you want to use are checked in the bottom list. After choosing OK, recording sources can be selected and adjusted. For information about using or installing Volume Controls, refer to Windows help. Recording volumes can be set by using the Device Controls Volume Properties as well.

Device controls
Use this command to show or hide the Device Controls window. See the Device Controls Overview section for more information.

CD audio extraction
The CD Audio Extraction tool digitally copies audio directly from an audio CD to a file on your hard drive, without using your sound card. This features has several advantages over normal recording:

There is no need to create and initialize a new file. Audio can be compressed while extracting, saving lots of hard drive space. System noise and interference is eliminated. Recording volume controls do not have to be selected or adjusted. Synchronizing recording with CD playback is unnecessary. Some CD-ROM drives allow extraction several times faster than recording.

Note that only recent SCSI and ATAPI (IDE) CD-ROM devices are supported. For Windows 95 and 98 you must have an ASPI driver installed. Due to the wide variety of interfaces and inconsistent device standards, incompatiblities may arise that will require a system reset. It is recommended that you close all other programs before proceeding. If your system is configured correctly, you will see a dialog that allows you to select a CD device and specify the region of audio to copy. If you have only one CD-ROM drive, only one device will be listed in the drop down list. If you already have an audio CD in the drive, the track times will be shown in the From and To time lists. Otherwise you will need to insert an audio CD and select the device from the list. The Preview from button plays the first five seconds beginning at the From time. The Preview to button plays the last five seconds up to the To time. Choosing the Save button will prompt you for a filename and format and then begin extraction. The Read technique group lets you manually specify what SCSI command to use to read from your CD-ROM drive. In some cases, GoldWave will automatically select an appropriate technique. In general, you will have to try each option to see which one works on your system. If you have an ATAPI (IDE) CD-ROM drive, start with the MMC option. The Fix defects option helps to eliminate pops and clicks caused by slight reading alignment errors (also know as jitter). When enabled, the CD is read more than once to detect and correct any errors. If your CD-ROM drive automatically fixes these errors (such as Plextor or CD-R devices), you can disable this option to speed up extraction. When this option is enabled, you may receive synchronization errors. If such an error occurs at the beginning, you may have to continue many times before synchronization is reliably established. The Byte swap option changes the order of bytes (endian) extracted from the CD-ROM. If the audio you extracted is badly distorted or a loud hiss, you will need to check or uncheck this box.

Troubleshooting If you get synchronization errors at the beginning of extraction, try continuing until extraction progresses normally. If the errors occur in the middle of extraction, the disk may be scratched or may need to be cleaned. If you see an error message indicating a read problem, make sure that the CD is free of dust and finger prints. If the CD appears clean, try using a different read technique. If you have tried all of the techniques and still get an error message, then CD audio extraction may not be supported on your system.

Options Menu Commands


Colours
Use this command to change the colour scheme of Sound windows. The Preview box shows the Sound's current colour scheme. The Colour Settings group allow you to make changes. To set the colour of a particular item, select the item from the drop-down list or click on the item in the Preview box. Once the item is selected, you can change its colour by clicking on one of the colour boxes.

Controls
Sets the Device Controls Properties.

File
The File Options dialog (Figure 20) lets you setup folders and file storage options. This sections assumes you are familiar with the terms introduced in the Storage Overview section.

Figure 20: File Options

Sound files This specifies the folder where you keep your sound files. The File | Open command automatically lists files in this folder whenever you start GoldWave. If you prefer to use the Windows 95 Properties feature to specify a working folder, enter a period, ".", for this folder. Temporary This specifies the folder to use when creating temporary files. This folder should be located on a large disk with plenty of free space. Using a compressed drive is not recommended. It will slow processing and give poor results when recording if hard disk storage is enabled. Changing this folder does not affect opened files already in temporary storage. Undo Undo specifies the folder to use for storing undo data. In most cases, it should be the same as the temporary folder. Changing the undo folder does not affect the current session of GoldWave since the undo file will have been created already. Undo can be enable or disabled by checking or unchecking the check box. Flash open The Flash open radio buttons let you control the flash feature. If you usually edit small files or have a slow system, set this to Never. If you always play files and rarely modify them, choose Always. If you have a fast system and often modify files, choose Limit and specify the minimum size (in units of 1000 samples) for a file to be flash. Any file larger than this will be flashed. Temporary storage This specifies where files should be stored for processing. RAM storage is usually very fast, but limits the size of files. Hard disk is slower, but allow huge files to be processed. Changing this options does not affect files already opened. Clipboard Due to size limitations in the standard Windows clipboard, copying large sounds can cause some problems. If you encounter any unusual behaviour or you frequently work with large files (20 megabytes or more), choose GoldWave to use GoldWave's special clipboard. If you edit small files and need to copy and paste between different applications, choose Windows to use the standard clipboard. The standard Windows clipboard may freeze your system when copying or pasting large audio clips. If this happens, use the GoldWave clipboard instead.

File types
Use the File Types dialog (Figure 21) to associate a filename extension (such as .snd or .vox) with an audio format. This is useful for automatically opening files that do not contain any information describing their format. For example, if you work with Dialogic files, you can associate the .vox extension with one of the Telephony formats. Whenever you open a .vox file, GoldWave will assume the format you specified and open the file without prompting you for the format.

Figure 21: Associating File Types To associate .vox with Dialogic ADPCM encoding: 1. 2. 3. 4. 5. Enter the extension "vox" in the box. Do not enter the leading period. Select 8000 from the Rate drop down list. Select Telephony from the Format drop down list. Select "4-bit VOX ADPCM, mono" from the Attributes drop down list. Choose the [+] button.

To remove an association: 1. Select the extension from the drop down list. 2. Choose the [-] button. To change an association: 1. Remove the association, as above. 2. Add the changed association, as above. The OS Associate button makes Windows associate GoldWave with the given file type so that double-clicking on a file in Windows Explorer opens the file in GoldWave. OS Association also adds Open and Play commands to the context menu for the file type when you right-click on a file. You might want to add an association for .wav files, since they are most common. Simply enter "wav" in the box and press the OS Association button. To associate wave (.wav) files with GoldWave: 1. Enter the extension "wav" in the box. Do not enter the leading period. 2. Choose the OS Associate button.

Tool bar
The Tool Bar Options dialog (Figure 22) lets you customize the tool bars in GoldWave. The two windows show all of the buttons available for the tool bars. To add or remove a button, simply find the button in the Main bar or Effects bar window and click on it. Buttons that are grayed will not appear in a tool bar.

Figure 22: Tool Bar Options

Window
Use the Window Options dialog (Figure 23) to configure the positions of the Main window, Sound windows, set axis options, and specify the zoom value for View | User.

Figure 23: Window Options Main window size This controls the Main window's position and size when GoldWave is started. Normal gives control to Windows. Maximize makes the Main window occupy the entire screen. Save position saves the Main window's position and size when GoldWave is closed so that it will appear in the same location next time. Sound window size This controls the position and size of Sound windows. Normal gives control to Windows, which usually results in cascaded windows. Maximize makes a Sound window occupy the entire Main window. Auto-tile resizes all Sound windows whenever a new sound is opened or closed so that every one will be visible.

Amplitude axis Displays or sets the units of the vertical axis in Sound windows. Selecting Off hides the axis completely. Normalized shows an axis with a range of -1.0 to 1.0. Signed 16-bit shows an axis ranging from -32768 to 32767, which is the range of a 16-bit sample. Unsigned 8-bit shows an axis with a range of 0 to 255. Time axis Sets the format for displaying the horizontal time axis in Sound windows. Seconds gives the time as a floating point number, such as 1234.567. Minutes gives the time as minutes:seconds, such as 12:34.567. User zoom This is where you specify the level of zoom for View | User and the User button. Values between 0.01 to 1000 are valid. Smaller values show more detail. Larger values show more of the sound.

Window Menu Commands


These commands organize Sound windows, tool bars, and status bars. Tile arranges Sound windows side-by-side so that they are all fully visible. Cascade layers Sound windows on top of each other so that their title bars are visible. Arrange icons arranges minimized Sound window in rows on the bottom of the Main window. Close all closes all Sound windows. You will be asked to save any sounds that have been modified. The Companion submenu lets you hide or show tool bars or status bars. A list of all currently opened Sound windows is given at the bottom of the Window menu.

Help Menu Commands


Contents starts Window's Help and gives a list of contents for GoldWave help. Using help provides instructions for using Window's Help utility. About displays version and registration information. The amount of available virtual memory is shown under the GoldWave icon.

Appendix A: An Introduction to Digital Audio


Digital Audio Attributes
Digital audio is composed of thousands of pieces of data, called samples. Each sample holds the loudness, or amplitude, of a sound at a given instant in time. This is similar to computer graphics where each point of light (pixel) has a certain brightness and location. All these points combine to make a picture. In digital audio, all the samples combine to make a sound. There are several attributes that determine the quality and quantity of digital sound. They are the sampling rate, the number of bits, and the number of channels. The Sampling rate is the number of times, per second, that the amplitude level is recorded. It is measured in Hertz (seconds-1, Hz). A high sampling rate results in high quality digital sound in the same way that high graphics resolution shows better picture quality. Compact disks, for example, use a sampling rate of 44100 Hz, whereas telephone systems use a rate of only 8000 Hz. The rate to use depends upon the type of sound and the amount of memory and disk space you have available on you system. Higher rates consume a lot of space. In the above example, the compact disk requires over 5 times the amount of storage as the telephone system for the same digital sound. Certain types of sounds can be recorded at lower rates without loss of quality. Some standard rates are listed in Table A.1. The number of bits determines how accurately the amplitude of a sample is recorded. The two most common are 8 bit and 16 bit formats. In an 8 bit sample, there are 256 different levels of amplitude. 16 bit samples have 65,536 levels. To compare the difference, let's say that you are a teacher grading tests and you can use one of two marking schemes (Figure 24). In scheme #1, the mark is out of 10. In scheme #2, the mark is out of 1000. All marks must be rounded off (no decimals allowed). If a student gets two thirds of the questions right, then in scheme #1, the grade will be 7 out of 10. In scheme #2, the grade will be 667 out of 1000. Obviously, scheme #2 is much more accurate. In digital sound, low levels of accuracy can cause noise due to quantization errors, as discussed later. Table A.1: Sound attributes Attributes Quality and Sound type 11025 Hz 8 bit mono 11025 Hz 16 bit mono 22050 Hz 8 bit mono Fair quality. Good for speech and low pitch sounds. Speech, less noise. Bytes/Second 11025 Bytes/Minute 662000

22050

1323000

Good quality. Good for music and relatively complex sounds.

22050

1323000

22050 Hz 16 bit stereo 44100 Hz 16 bit mono 44100 Hz 16 bit stereo

Very good quality stereo. Less noise. Excellent quality. Good for all sounds. Excellent quality stereo (CD quality). Large storage requirements.

88200

5292000

88200

5292000

176400

10584000

The number of bits is becoming less important due to the variety of audio compression techniques available today, such as RealAudio and MPEG. By using MPEG compression, for example, you can make a sound file anywhere from 10 to 100 times smaller and still maintain excellent quality. In compressed audio, the term "bits" has no meaning in the classical sense.

Figure 24: Bits Example

Digital audio can have one or more channels. Single channel audio, referred to as a monaural (or mono) audio, contains information for only one speaker and is similar to AM radio. Two channel audio, or stereo audio, contains data for two speakers, much like FM stereo. Stereo sounds can add depth, but they require twice as much storage and processing time as mono sounds. Movie theatres often have advanced audio systems with 4 or more channels, which are capable of making sounds appear to come from certain directions. Note that GoldWave supports mono and stereo sound only.

Byte Swapping and Sign


In a computer, 8 bit samples fit perfectly into a single byte. However, when more bits are involved, such as 12, 16, or 32 bit samples, more than one byte is required. Different processor designs (i.e. Motorola/Mac and Intel) store these bytes in a different order. Intel uses a "little endian" order, while Motorola uses a "big endian" order. For a sound created on one system to play correctly on another, the bytes have to be reorder. GoldWave designates this as byte swapped. All samples may be interpreted as signed or unsigned values. In general, 16 bit samples are always signed, so they are interpreted as values ranging from -32768 to -32767. The same is not

true for 8 bit samples. On a PC, 8 bit samples are usually unsigned, so values range from 0 to 255. On a Mac, they are usually signed, so values range from -128 to 127. This inconsistency makes it necessary to explicitly state whether a sample is signed or unsigned.

Frequency and Pitch


Sound windows in GoldWave show sound as a waveform of amplitudes on a time axis. However, sound can be viewed in an entirely different way by examining its frequency/pitch content or frequency spectrum. From this perspective, all sounds are formed from a combination of simple fundamental (sinusoidal) tones. A rainbow is an example of a frequency spectrum where visible light is broken down into bands of simple fundamental colours, such as red, yellow, green, etc. The actual colour of the sun is formed by combining these colours. A similar process applies to audio.

Frequency Ranges
Average human hearing spans a frequency range from about 20 Hz to about 17000 Hz. Figure 25 shows some common sounds and the frequency range they cover.

Figure 25: Frequency Range of Sounds Many people wonder why it is difficult to remove vocals from music. From Figure 25, we see there is a large overlap in the frequency range of speech and music. Removing the vocals would also remove a significant part of the music. A similar problem occurs when removing hiss noise, since it often covers the entire spectrum. Most basic stereo systems have bass and treble controls, which offer limited control over a frequency spectrum. Bass applies to low frequency sounds, such as drums, cellos, low piano notes, or a hum noise. Treble applies to high frequency sounds, such as a clash of cymbals, a tweet of a small bird, high notes on a piano, or a hiss noise. More expensive stereo systems have Graphic Equalizers, which provide better control over a frequency spectrum. Instead of controlling just two bands (bass and treble), you can control many bands.

GoldWave provides even more control over frequency spectrums with filter effects such as Parametric EQ, Highpass, Lowpass, Bandpass, Bandstop, and Equalizer.

Frequency Range and Sampling Rate


The frequency range of a digital sound is limited by its sampling rate. In other words, a sound sampled at 8000 Hz cannot record frequencies above 8000 Hz. In fact, the sound cannot even have frequencies above 4000 Hz. According to the sampling theorem, the maximum frequency is limited to half the sampling rate. CD audio is designed to cover the full range of human hearing, which has a maximum of just under 20 kHz. In order to successfully record this range, the sampling theorem states that we must use a sampling rate at least twice that maximum. That is why CD audio uses a sampling rate in excess of 40 kHz.

Frequency Spectrum Graphs


Several of GoldWave's Device Controls graphs transform sounds into a range of frequency bands using a radix-2 fast Fourier transform (FFT) algorithm. When the results are drawn using colours, the graph is referred to as a spectrogram. When the results are drawn with lines, it is often referred to as a frequency spectrum or frequency analysis. Frequency analysis graphs are displayed in the Noise reduction and Parametric EQ filter effects. These help you to locate frequencies that you would like to remove or enhance. GoldWave optionally applies a windowing function to the data before performing the FFT (see Device Controls Graph Properties section). This reduces "discontinuity" errors that occur when dividing data into small chunks. The Hamming window is used by default. The Kaiser window also gives good results. To make the spectrum more realistic to human hearing, magnitudes are scaled logarithmically. This means that if one frequency "sounds" twice as loud as another, it will be graphed with twice the height (or the corresponding colour for the spectrogram).

Problems with Recording


There are five potential problems when recording sound: aliasing, clipping, quantization, internal noise, and system configuration. Aliasing occurs when the sampling process does not get enough data to correctly determine the shape of the sound wave. The recorded sound will have missing tones (Figure 26, top) or new tones that never existed in the original sound (Figure 26, bottom). These problem can be eliminated by using higher sampling rates or by using anti-aliasing filters.

Figure 26: Aliasing Higher sampling rates increase the number of sampling points. To see how this works, try adding a few points between each sampling point in the figure and redraw the graph. The recorded sound will more closely resemble the input. Anti-aliasing filters remove all tones that cannot be sampled correctly. They prevent high pitched tones from being aliased to low pitch. Many sound cards include anti-aliasing filters in hardware. Clipping errors occur when the sampled amplitude is outside the range of valid values. If, for example, the range is -1.0 to 1.0, and a value of 1.2 is sampled, then the value must be clipped to 1.0 (see Figure 27). This generates distortion. To eliminate clipping, adjust the recording volume before recording. By using the Device Controls' monitor feature, you can analyse the level to determine a suitable volume. The volume is low enough when the red LEDs remain off.

Figure 27: Clipping Quantization errors occur when the sample is rounded to the nearest level of amplitude. This can be explained by using the "marking schemes" example in the previous section. The number two thirds (2/3) is represented by 7/10 in scheme #1. This gives a quantization error of: | 7/10 - 2/3 | = 1/30 Similarly, in scheme #2, the quantization error is: | 667/1000 - 2/3 | = 1/3000 Clearly, scheme #2 has the smallest error. Therefore, using 16 bits instead of 8 bits is a good way to reduce quantization errors.

The other two recording problems deal with computer hardware and software. To minimize internal noise, make sure your audio card is installed as far away from your graphics card as possible. If you use a microphone, keep it away from your monitor and computer fan. Remember to use shielded cables. System configuration can also affect audio quality. Recording to a compressed drive (DriveSpace) is not recommend. Compression ratios on audio are generally poor and the CPU overhead can cause gaps during recording. Due to architectural problems with PCs and excessive virtual memory swapping by Windows, you may notices an occasional gap when recording. If such a gap occurs at the beginning of recording, disable the Allow undo option under Device Controls Recording Properties.

Appendix B: Keyboard Commands


In addition to all the standard menu keystrokes, such as Alt+F O to open a file, Alt+E C to copy, Alt+E V to paste, etc., GoldWave includes a number of additional quick keystrokes. These are summarized in the following table. Note that most keystrokes work only when the Main window is active and not when the Device Controls window is active. Keystroke Ctrl+A Ctrl+B Ctrl+C or Ctrl+Ins Ctrl+E Ctrl+F Ctrl+J Ctrl+Shift+J Ctrl+M or Shift+Ctrl+Ins Ctrl+N Ctrl+O Ctrl+P Ctrl+Q Ctrl+R Ctrl+S Ctrl+T Ctrl+V or Shift+Ins Ctrl+W Ctrl+X or Shift+Del Ctrl+Z or Alt+Backspace Shift+A Action Selects the entire sound. Pastes clipboard into the sound at the beginning. Copies selection into the clipboard. Pastes clipboard into the sound at the end. Pastes clipboard into the sound at the finish marker's position. Jumps the start marker to the next cue point. Jumps the start marker to the previous cue point. Mixes clipboard with the sound at the start marker's position. Creates a new sound. Opens a sound. Pastes clipboard into a new Sound window. Drops a new cue point at the current playback or recording position. Replaces the selection with the clipboard contents. Saves the file. Trims the sound. Removes all audio outside the selection. Pastes clipboard into the sound at the start marker's position. Sets the select to the view (as in Select View). Removes selection and copies it into the clipboard. Undoes last change. Horizontally zooms all the way out.

Shift+P Shift+S Shift+U Shift+1 Shift+V Shift+Tab Tab Del [ (left bracket) ] (right bracket) Left Right Pg Up Pg Dn Home End Ctrl+Home Ctrl+End Shift+Left, Shift+Right

Zooms to previous horizontal zoom. Horizontally zooms in on the selection. Horizontally zooms to the user defined level. Zooms 1:1 horizontally. Vertically zooms all the way out. Stores the locations of the start and finish markers. Moves the start and finish markers to the stored locations. Deletes the selection, permanently. Drops the start marker at the current playback position. Drops the finish marker at the current playback position. Scrolls the Sound window view left. Scrolls the Sound window view right. Scrolls the Sound window view left one page. Scrolls the Sound window view right one page. Moves the Sound window view to the start marker's position. Moves the Sound window view to the finish marker's position. Moves the Sound window view to the beginning of the sound. Moves the Sound window view to the end of the sound. Moves the start marker left or right.

Ctrl+Shift+Left, Moves the finish marker left or right. Ctrl+Shift+Right Shift+Up Shift+Down Ctrl+Up Ctrl+Down Space Horizontally zooms in. Horizontally zooms out. Vertically zooms in. Vertically zooms out. Plays a sound (when Main window is active).

Plays/Stops a sound (when Device Controls window is active). F4, F5, F6, F7, F8 Shift+F4 Plays, rewinds, pauses, fast forwards, and stops respectively. Plays the sound with the User play button settings.

Ctrl+F9, Ctrl+F8 Starts and stops recording respectively. Pause Esc Scroll Lock F11 F1 Ctrl+F4 Alt+F6 Ctrl+F6 Pauses a sound. Stops a sound. Scrolls the view during playback so that the current position is always visible. Displays Device Controls Properties window. Starts help. Closes the Sound window. Switches between Main window and Device Controls window. Switches between Sound windows.

Appendix C: Expression Evaluator


OVERVIEW The Expression Evaluator is a versatile tool for manipulating and generating audio data. After you select Expression evaluator from the Tools menu, you are presented with a dialog that is similar in appearance to a calculator. The Destination is the Sound window where results of the evaluation will be stored. The drop down list contains all Sound windows in the form "X - NAME", where X is the wave identifier number of the Sound window and NAME is the filename of the sound. For example, a Sound window with the title "HELLO.SND" could appear as "1 - HELLO.SND" in the list. By default, the destination is set to the current Sound window. You can change the destination, if more than one Sound window is opened, by using the up and down keys or by selecting it with the mouse from the drop down list. A Source is a Sound window containing existing audio data that will be used in the expression. By selecting a source from this list, the function waveX( will be placed in the expression. X is the wave identifer number, as explained above. A large Expression box is located in the middle of the dialog. This is where an expression is entered. A list of valid operations and functions is given in a following section. In most cases, expressions will be some function of n or t, just as in regular math, where y is usually a function of x (i.e. y = f(x)). In the expression evaluator, we can have destination = f(t), where f(t) is any expression you enter. To create a simple tone, for example, you would enter the expression sin(600*t). You can even alter an existing sound. To double the volume of "1 - HELLO.SND", for example, you would select it as the destination and enter the expression wave1(n)*2. To enter an expression, you can: 1. Type it in using the keyboard. 2. Click on the "calculator" buttons with the mouse. or 1. Use the Group name and Expression name boxes to select a previously saved expression. The evaluator uses three special variables, which you can initialize in the Variables box. These variables are discussed later. After you have specified the destination, expression, and initial values, choose the Start button (or just press the "Enter" key) to begin evaluation. If you entered an expression incorrectly, a message will be displayed by the word Status. The Status area also gives the progress during evaluation. Since the evaluation process takes time, you can stop it at any time with the Cancel button. Pressing the Cancel button a second time will close the Expression Evaluator dialog. You can copy, cut and paste expression in the Expression box using the usual keystrokes (Copy = Ctrl+C, Cut = Ctrl+X, Paste = Ctrl+V). You can also copy and paste expression from the online help.

To speed up evaluation, make sure that you are using RAM storage (see Options | File). A co-processor can dramatically increase the speed of the evaluation, since the evaluator uses floating point calculations. By creating a new mono sound with a low sampling rate (11025 Hz), you can experiment with expressions quicker. You can then create a sound with a higher sampling rate for the final evaluation.

Evaluation Range, Variables, and Constants


Knowledge of the structure of digital audio is essential to understand how the evaluator works. To illustrate this structure, let's assume we have the following sound:
Title bar: Total length: Sampling rate: Start marker: Finish marker: HELLO.SND 2.0 seconds 8000 Hz 0.5 seconds 1.2 seconds

Figure 28: Internal Structure of HELLO.SND, 8000 Hz, 2 Seconds Digital audio is stored as a series of amplitudes, which are often referred to as samples (see Figure 28). The evaluator interprets each sample as a value between -1 and 1, inclusive. If the result of an evaluation is outside this range, it will be clipped. Only samples between the start and finish markers are considered valid; all other values are assumed to be zero. The number of samples selected is defined as N. Each sample has a relative index number, n, and a time, t. Since the time of each sample depends on the sampling rate, it is usually written in terms of the unit of time between each sample, T. You many have noticed that the time, t, is related to the index number, n, by the equation t=nT. Figure 28 shows how all these variables relate to the structure of the sound.

Using Time in an Expression


Let's assume we have entered the expression sin(t). Since expressions are evaluated over the selection range, the initial value for t is automatically set to start marker's position of 0.5. By choosing the Start button, the expression will be evaluated from t = 0.5 to t = 1.2 in steps of 1/8000 of a second, as defined by T. This means that the expression is calculated for each sample in the selection, changing each sample as follows: Sample4000 = sin(0.500000) Sample4001 = sin(0.500125) Sample4002 = sin(0.500250) ... Sample9600 = sin(1.200000)

Using the Sample Index in an Expression


The sample index is useful for modifying an existing sound. If we want to double the amplitude of HELLO.SND, for example, we need to multiply each sample by two and store it back into the sound. In this case, HELLO.SND will be both the destination and the source. To set it as the destination, we simply select it from the Destination list. To use it as a source, we need to determine its wave identifier number. These numbers are provided in the Source list. Assuming it is listed as "3 - HELLO.SND", we now know that its wave identifier number is 3. This number is necessary for the evaluator's wave function, which has the following syntax: waveX(n) where: X = wave identifer number n = sample index number In the evaluator, the index number, n, is relative to the start marker. This means that the start markers position is added to the index number (i.e. n+Start). For the example in Figure 28, a relative index of n=0 has an absolute index of 4000. The Start sample always has a relative index of 0. The final expression is wave3(n)*2. By choosing Start, this expression will be evaluated from n=0 to n=5600 in steps of 1 (note that N = 5600) . This produces the following changes (remember than n is relative): Sample4000 = Sample4000 * 2 Sample4001 = Sample4001 * 2 Sample4002 = Sample4001 * 2 ... Sample9600 = Sample9600 * 2

Note that N and n are always integers. The evaluator rounds indices to the nearest integer, so the expression wave3(.7) would be calculated as wave3(1). You can use the sample index number and the wave function to mix two or more wave together. If you have several sounds opened, you can obtain the wave identifier number for each sound from the Source list. If the sounds you wanted to mix were identified as 2 and 3, you would enter the expression: wave2(n) + wave3(n) Care must be taken when indexing signals with different sampling rates. Assume wave1 is a voice recorded at 11025 Hz and wave2 is music recorded at 22050 Hz. If you want to mix these two signals, with wave1 as the destination, then the expression wave1(n) + wave2(n*2) must be used (ideally, wave2 would have to be low-pass filtered first). Whereas, if wave2 is the destination, the expression should be wave1(n/2) + wave2(n)

A variable N has several uses, such as reversing a sample. Assume wave2 is a new sound that has the same sampling rate and length of wave1. By setting the destination to wave2 and using the expression wave1(N-n) wave2 will be the reverse of wave1.

User Variable f
The user variable, f, can be set to any value you choose. In many cases, this value is used as a frequency, hence the letter "f". For example, if you entered the expression sin(2*pi*f*t) you can then generate any sine wave by specifying the frequency in the f box. This value does not change during evaluation, but stays at the value you assign to it. Technically, one might classify f as a constant.

Conversion Between Variables


The following equations convert between time and sample index number. The start parameter is the position of the start marker (in seconds). n = (t - start) / T t = nT + start T = 1 / (sampling rate)

Group name and Expression name


The Group name and Expression name boxes allow you to organize and store expressions in the express.eqx file located in your Windows folder. Similar expressions can be stored together in groups. The Group name box lists all of these groups, while the Expression name box lists all the expressions in a group. To retrieve an expression: 1. Select the group from the Group name list. 2. Select the expression from the Expression name list. To add an expression: 1. 2. 3. 4. Enter the expression. Type in the group name or select a group from the Group name list. Type in an expression name. Choose the Add button.

To delete an expression: 1. Select the group from the Group name list. 2. Select the expression from the Expression name list. 3. Choose the Delete button.

When a group becomes empty, it is deleted automatically.

Evaluator Operators and Functions


The following table summarizes evaluator operators and functions. Table B.1: Evaluator Operators and Functions Label (, ) +, *, -, / % ^ pi cos sin tan acos asin atan cosh sinh tanh sqrt abs sgn log, ln exp step int rand(n) wavex(n) Operation, function Parenthesis Add, multiply, subtract (negate), and divide Modulus operator (remainder) To the power of, yx Constant (3.14159...) Cosine Sine Tangent Arccosine Arcsine Arctangent Hyperbolic cosine Hyperbolic sine Hyperbolic tangent Square root Absolute value Sign (-1, 0, or 1). Log base 10, natural logarithm Exponential base e Unit step ( 0 for t < 0, 1 for t >= 0 ) Integer value Random number between 0 and n Sound amplitude at n. x specifies the Sound window as given in the Source list. If no x is specified, the destination Sound window data is used.

Signal Generation
Several signal generation expressions are listed below. Words given in italics represent numeric values that you must enter. To try one of the following expression, perform the following steps: 1. 2. 3. 4. Choose New from the File menu. Choose OK. Choose Expression evaluator from the Tools menu. Type in the expression as given in the example. For example:
sin(2*pi*261.7*t)

5. Choose Start. 6. Wait for the "Finished" message. 7. Play the sound. Note that you can play the file during evaluation. Table B.2: Expressions Type General Expression Examples Middle C: sin(2*pi*261.7*t) Sine wave sin(2*pi*frequency*t) Telephone dial tone for "5": (sin(2*pi*1336*t) + sin(2*pi*773*t)) / 2 200 Hz tone: 1 - 2*abs(1 - 2*200*t%2) Full volume white noise: 1 - rand(2) 400 Hz tone: int(2*t*400)%2*2-1 Slow sweep up to 20 kHz: sin( 2*pi*160*(t%5)^3 ) 50% decay a 500 Hz sine wave: (0.5*exp(-t) + 0.5) * sin(2*pi*500*t)

Saw wave White noise Square wave Sweep

1 - 2*abs(1 2*frequency*t%2) amplitude rand(2*amplitude) int(2*t*frequency)%2*2-1 sin(2*pi*t^rate)

Exponential (1 - minimum)*exp(-t) + decay minimum

Custom Filters
One of the easiest ways to create your own digital filters is to use Matlab (The Student Edition of Matlab, by The Math Works Inc., published by Prentice-Hall, ISBN 0-13-855974-0). It has many built-in commands that generate filter coefficients. The coefficients can then used in the Expression Evaluator or the User defined filter command.

Example of a Low-Pass Filter In preparation for down-sampling, you can use Matlab to generate the coefficients of a 4th order Butterworth low pass filter that will remove noise above half the Nyquist frequency ( the sampling rate). Enter: [b,a] = butter(4, 0.5) The result should be similar to:
b = a = 0.0940 1.0000 0.3759 0.0000 0.5639 0.4680 0.3759 0.0000 0.0940 0.0177

To implement this filter in the evaluator, let's assume that the sound to be filtered is in the Sound window titled SOUND.SND. 1. Use File | New to create a new Sound window with the same sampling rate as SOUND.SND. 2. Make sure the length of the new sound is as long (or longer) than the original. If it is not, use Edit | Insert space to increase its length. 3. Use Tools | Expression evaluator to open the expression evaluator dialog. 4. Set the destination to the new sound. 5. Enter the following expression (assuming SOUND.SND has a wave identifier of 1):
6. 7. 8. 9. wave1(n)*0.0940 + wave1(n-1)*0.3759 + wave1(n-2)*0.5639 + wave1(n-3)*0.3759 + wave1(n-4)*0.0940 - wave2(n-2)*0.4860 wave2(n-4)*0.0177

10. Choose Start.

Appendix D: Troubleshooting and Q&A


Troubleshooting
Problem Cannot open large files Cause/Solution Make sure that hard disk storage is enabled in Options | File. Make sure that you have plenty of free RAM and hard disk space. CD quality sound requires 10MB per minute and 30MB per minute when editing. Cannot play sounds GoldWave or audio device/driver is incorrectly installed. Make sure an audio driver is installed in the Device Manager or use the Control Panel "Add New Hardware" option to install a driver. Check that the Windows Sound Recorder accessory can play sounds. If it doesn't, the driver is not installed correctly. Make sure that your audio device is selected by using the Device Properties tab in GoldWave's Device Controls Properties. Cannot record sounds See above. Make sure you have the correct recording source selected. See Recording Sounds under Device Controls Overview. Make sure your audio device/driver can record sounds. Sound may be in use by the playback device; click on the stop button. Cannot use the stop or pause button or real-time graphs do not work The audio driver is synchronous. This means that Windows (and GoldWave) loses control until the sound has finished playing. Your system may not be fast enough to draw the real-time graphs. Try reducing the frames/s rate in Device Controls Graph Properties or resize the Device Controls window so that the graphs are hidden. System freezes or crashes or a General Protection Fault occurs Make sure that you have a 486 or better system. A Pentium system is recommended. If the freeze occurred after copying a large file, choose "GoldWave" as the clipboard under Options | File. If the crash occurs during recording or playback, install an updated sound card driver. You may have encountered a problem. If you can duplicate the problem, contact the author for more information. After a crash, there is less free space on my hard disk Graphs / LEDs are out of synch Delete files in the temporary storage folder specified under Options | File. These files usually have names like GWABCD.TMP. Note that you may be able to recover the files by specifying the format manually as "16-bit, stereo, signed" or 16-bit, mono, signed". Try using a different positioning method in Device Controls Device Properties. Many audio drivers return inaccurate "current" positions. Make sure you have the most up-to-date device driver. Periodic popping or Some old audio devices/drivers make periodic pop/clicks between DMA transfers

clicking

and/or memory boundaries. It is most noticeable when playing pure sine waves. If a newly recorded sound has pops or clicks at the beginning, disable the Allow undo feature in Device Controls Record Properties. Pops and clicks can occur at the beginning or ending of a sound if the first or last sample is not 0 (silence). Fading in/out a small selection can sometimes fix this. Enable the Edit | Marker | Snap to zero-crossing feature to avoid pops and clicks when editing.

Expression Evaluator slow

Make sure that RAM storage is selected in Options | File. Remember to close and reopen your sounds for this setting to apply. Your system does not have a co-processor.

Editing seems to be getting slower and disk activity is increasing Distortion in recording

Files on your hard disk are becoming fragmented. Use the Windows Disk Defragmenter system tool. You need to lower the recording volume. See Tools | Volume control for information about selecting and controlling recording volume. Check that all connections are correct and firm (do not connect a "line-out" to "micin", for example).

Gaps in recording and playback

The Windows virtual memory manager can sometimes cause gaps due to excessive swapping. If you notice gaps at the beginning of a recording, disable the Allow undo feature in Device Controls Record Properties. Your system may be too slow to display the real-time graphs. Resize the Device Controls window so that the graphs are hidden.

Sound won't play for more than a few seconds.

Make sure your driver is configured to play for more than few seconds. The old PCSpeaker driver, for example, will play for only 4 seconds unless you configure it to play longer.