TwistedWave
Reference Manual

1 Introduction

The TwistedWave reference manual is available both in a browsable format, from the Help menu, and as a printable PDF here.

What makes TwistedWave unique is its ability to deal with very large files. You don’t have to wait until TwistedWave has finished loading a file or applying an effect before you can start playing the result, or applying another effect. You can read more about this in section 2.3.

2 Getting running

The first thing you will do with TwistedWave will be to load an existing audio file, or create an empty document. These tasks are described in the two following sections.

2.1 Creating a new document

Create a new document by selecting File / New from the main menu, and you will be asked for the bit depth, sample rate and number of channels for the new document.

New document settings

The bit depth indicates the resolution at which the file will be saved. With a higher bit depth, the sound quality will be better, but the files will be larger. For good results, a value of at least 16 bits is recommended. Note that you can still change your mind, and select a different bit depth when you save the file. If you save in a compressed file format, such as aac or mp3, the bit depth will not have any influence on the resulting file.

For the sample rate, however, it is important to select the correct value from the beginning. It is possible to change the sampling rate afterward, but the conversion is never perfect, and will result in a slight reduction of the audio quality.

The third parameter is the number of channels. Select 1 for mono files, 2 for stereo files, or any other number you may need.

2.2 Loading a file

More commonly, instead of starting with an empty document, you will want to open an already existing file, and start working on it.

TwistedWave can handle many different file formats, and opening an audio file works similarly to any other application, by selecting File / Open... in the main menu.

TwistedWave also supports many movie file formats, and selecting one will extract its sound track.

2.3 Large files

When you open a very large file in TwistedWave, it can take a while for it to load completely. However, you can still immediately start playing the file, adding effects, copying and pasting parts of it or do whatever while the file is still being loaded!

It is only when you want to save the resulting file that you will have to wait for all the pending tasks to be completed.

When a task has not finished processing, the part of the waveform that is not up to date appears in a lighter color, and a progress bar at the bottom of the window indicates the effect being applied, and the amount done.

3 Manipulating the Waveform

As it is important to be able to quickly navigate in the sound file to find the portions you want to listen to, or edit, TwistedWave offers a number of ways to manipulate the waveform. Both zooming and panning are fast and easy. Here are the different ways to perform these.

3.1 Zooming

3.1.1 From the toolbar

This is the simplest and most obvious way to zoom. Two buttons in the toolbar allow you to easily zoom the waveform in and out.

3.1.2 With the scroll wheel

One easy way to zoom in the waveform is to use the mouse wheel. You can zoom in and out by scrolling up or down.

By default, when zooming in and out, TwistedWave makes sure the cursor remains in the center of the window. An option in the preferences window (section 18.2.1) can make this zoom be centered on the mouse pointer instead.

When this option is enabled, you can move the mouse pointer left or right to adjust the point where you want to go as you are zooming in.

3.1.3 By dragging

Another easy and very fast way to zoom in the waveform is to press the command key, and drag the mouse vertically. Move the mouse down do zoom in, and up to zoom out. You can also move the mouse horizontally to indicate where you want to zoom in.

This is a very powerful way to navigate in the waveform, because it allows you to zoom and scroll at the same time. In a single drag, you can zoom out, move the mouse over another location in the file, and zoom back in.

3.2 Vertical Zoom

When the sound file you are working on is very quiet, and the waveform does not show enough detail, you can zoom vertically to work more comfortably. TwistedWave offers three ways to do this:

With the Vertical Zoom toolbar icon. By clicking and dragging on this icon, you can adjust the vertical zoom. A single click will restore the zoom to its original value.
By click and dragging on the vertical ruler at the left of the waveform. This works the same as the toolbar icon. A double click will restore the zoom to its original value.
By using the scroll wheel with the command modifier key.

3.3 Scrolling

3.3.1 With the scroll wheel

If you have a mighty mouse, or another kind of mouse that offers a horizontal scrolling mechanism, you can use it to scroll the waveform in TwistedWave.

With a regular scroll wheel, you can also scroll horizontally by pressing the shift modifier key.

3.3.2 By dragging

Another way to scroll the waveform is to drag it with the mouse while pressing the control key.

Additionally, when you are selecting, TwistedWave will automatically scroll the waveform when the mouse reaches the edges of the document window.

3.4 The temporary zoom

3.4.1 Zooming out

As the pattern described above to move in the waveform (zoom out, change position, zoom in) turns out to be quite useful, there is a keyboard shortcut just for it. Press Shift-Z, and the wave will zoom out. When you release the Z key, it zooms back in at the same level you were before. When you keep the Z pressed, a window appears, showing you where you will be when you zoom back in. By moving the mouse horizontally, you can move the window to the location you want to be.

The zoom window

3.4.2 Zooming in

By pressing the Z key without the shift, you can perform the opposite operation. When you press it, TwistedWave will zoom in, and when you release it, it returns to the original position. This can be useful if you want to see more closely a small detail, without loosing your current position. What makes this feature very interesting is that when you are zoomed in, you can still manipulate the wave with the mouse. If you want to select an area with great precision you can do the following:

Press Z to zoom where you want to start the selection
Click to start selecting
Without releasing the mouse, release Z to zoom back out
Press Z again at the location where you want to end the selection
Release the mouse to end the selection
Release Z to zoom back out

Note also that when you are zoomed in, you can continue zooming or panning with the scroll wheel, the scrollbar, or by manipulating the red window in the overview, control dragging or whatever. When you release the Z, the view will always revert to the position you were before you started zooming. This can be useful if you want to see or select something at another location with the ability to return back where you were just by releasing the Z key.

3.5 Using the overview

The overview is the area above the main waveform display that shows the whole file, with a red window indicating the portion visible in the main display. This red window can also be manipulated in order to change the visible portion of the sound file. You can drag it to another position, or reshape it by dragging one of its edges in order to see a smaller or larger area.

If you press the option key while dragging the window, you will be able to make it shrink or grow, effectively zooming in or out of the waveform.

4 The selection

By default, all the effects available will be applied to the whole sound file. Often, you want an effect to be applied to a small portion of it. You can do this by selecting the area you want the effect to work on before applying the filter.

A document with a selected area

4.1 Selecting

In order to select a part of the sound file, just click and drag the mouse over the area you want to select. In order to get more precision on the area you are selecting you can temporarily zoom at the places where you want to start and stop selecting. This process is described in more detail in section 3.4.

It is possible to adjust the current selection by dragging one of its edges. This is done by pressing shift when clicking on the waveform.

4.2 Selecting whole channels

It is often necessary to select a whole channel. TwistedWave provides a convenient way to do this by double-clicking on a channel. All the data on that channel will be selected from the first to the last frame. After having double clicked, you can keep the mouse button down, and drag the mouse pointer over adjacent channels to select them as well.

By using the option or shift-option key combinations while double clicking, you can add or remove whole channels from the current selection.

By triple clicking, you can also select all the channels at once. This is equivalent to using the Select / All menu item.

4.3 Selection editor

A powerful way to edit the selection is with the selection editor. It can be opened from the menu Select / Edit Selection…, or simply by clicking on the selection length in the toolbar (The selection length is not present in the default toolbar. You can add it by right-clicking on the toolbar, and selecting Customize Toolbar….).

Selection editor

The selection editor allows you to enter numerical values for the begin, end and size of the selection. When you enter a time in any of these fields, TwistedWave understands many formats. Simply entering a numerical value such as 5 corresponds to a number of seconds. You can also enter 1m25 to indicate one minute and 25 seconds. 3h means 3 hours, 15" is 15 seconds, and 1’23 is one minute and 23 seconds.

On the right hand side of the selection editor, a number of check-boxes are used to indicate which channels have to be selected. This can be used to select only the left or right channel in a stereo file, for instance.

4.4 Cursor editor

Exactly as you can enter the numerical values for the selection, the cursor editor allows you to enter the time you want to move the cursor to.

The cursor editor is opened from the menu Select / Edit Cursor…, or by clicking on the cursor position in the toolbar.

Cursor editor

4.5 Advanced selection

Usually, a selection is just an area contained between two samples. With TwistedWave, you can have more complex selections that are composed of several disjoint areas. Such selections can be useful when you want to apply an effect at different places in the file. Instead of selecting an area, applying the effect, selecting another area, and reapplying the effect, you can select all the areas at once, and then preview the effect on them all, and apply it.

A complex selection

Two key combinations allow you to create complex selections. The first is by using the alt/option key. When it is pressed, the selections you make by dragging the mouse are added to the current selection. You can also remove an area from the selection by using both the shift and alt keys. As with simple selections, it is still possible to adjust the edges of complex selections with the shift key.

4.6 Auto extend to crossing

When the Auto Extend to Crossing option from the Select menu is enabled, the selection is automatically extended by a few samples to the next zero crossings of the wave at its edges. This is done in order to avoid discontinuities when the sample is cut and pasted somewhere else. A discontinuity in the sound wave would result in a click when played, and should be avoided. When you move the insertion cursor, it is also automatically moved next to the closest zero crossing. That way, when you paste something at that position, this will not introduce a discontinuity.

$The selection was automatically adjusted\n to start and end at a position where the wave crosses the zero\n axis.$

The selection was automatically adjusted to start and end at a position where the wave crosses the zero axis.

It is important to note that if TwistedWave has not finished loading the file you are working on, or if it is currently applying an effect, the Auto Extend to Crossing function will not work. The reason is that because in such a case, TwistedWave cannot accurately determine the position of the zero crossings. It is therefore not possible to adjust the selection accordingly.

4.7 Detecting silences

Imagine you have recorded a long file, and you would like to cut the silences, make them shorter, or split the different parts separated by silences. This can be a very time consuming task, and should certainly be automated.

From the main menu, click Select / Detect Silences..., and a silence detector will come to the rescue.

The silence detector

When the silence detector is open, it will automatically analyze the audio file, and detect the silences. In order to preview the silences detected, they will automatically be selected.

Previewing the silences

Alternatively, you can choose to select the sounds instead of the silences. That way, you can start playing the audio, and will hear all the regions that do not contain silences.

When the silences have correctly been analyzed and detected, you can perform on them any of the actions available from the pop-up button:

Select Silences. This option will select the parts of the audio containing silence.
Select Sounds. This is the opposite of the previous option, and only the areas not containing silence will be selected.
Place Markers. Markers will be automatically placed at the beginning and end of each detected silence.
Delete Silences. All the silences will be deleted.
Delete Silences and Mark. All the silences will be deleted, and some markers will be placed where the cuts were made.

The silences were cut, and markers were added

4.7.1 Expert silence detector

If the simple silence detector fails to detect the silences contained in the audio file, you can switch to the "Expert mode" where a number of parameters allow you to fine-tune the silence detection.

$The expert silence detector, with extra\n parameters$

The expert silence detector, with extra parameters

Threshold. This parameter specifies the sound level in dB below which a sound is considered as silence. Note that when you open the silence detector, the sound file is analyzed, and this parameter is automatically set to a value that should be fine. You should only have to change this parameter in some cases where the distinction between sound and silence is less obvious.
Minimum silence duration. This parameter specifies the minimum duration below which a silence is not considered. This can be useful when detecting silences from a file containing spoken text, and a small pause between two words should not be considered as silence.
Minimum sound duration. This parameter specifies the minimum duration below which a sound is not considered as such, and will be treated as a silence. When a short noise is present during a silence, it can be desirable to still consider the silence as a whole, and ignore the sounds up to a given length.
Left and right padding. When you want to cut the silences from a sound file, it may be necessary to keep around a fraction of a second of audio before and after each sound. If the sound fades out at the end, for instance, it may be necessary to keep it a bit longer even after it drops below the silence threshold.

4.7.2 Splitting a file

A common use for silence detection is splitting an audio file containing several sounds into several audio files. In order to do this, select the option Delete Silences and Mark when detecting silences.

You should then rename the markers to give them the names you would like to give to your files. You can easily do that with the markers window (see section 11.3).

And you can then export all the parts separated by markers from the main menu, Markers / Split by Markers (see section 11.4).

4.8 Detecting transients

One of the most difficult tasks when editing an audio document is to properly select the area you want to edit. The capture here shows TwistedWave with a file loaded, containing a percussion loop. The individual shots are clearly visible, but accurately selecting one of them requires quite some dexterity.

A selected transient

This is where the transient detection in TwistedWave comes handy. This allows TwistedWave to automatically detect the individual shots in the file. The operation Select / Move Cursor to Transient, available with the Tab key by default, moves the cursor to the next transient in the file, and can be used to easily move to the start of every individual shots in the file. You can then Shift-Tab to select the area between the cursor position and the next transient. This is how we could easily select a single shot, as shown in the capture here.

The transients in the example shown here are clearly identifiable. Often, however, there may be a lot of background noise, and the transients may not be as strong. If TwistedWave fails to detect the transients, you can adjust the detection threshold from the preferences window (see section 18.2.4).

4.9 Copy / pasting

Just as in a text editor, once you have selected an area of the sound file, you can copy the selection, and paste it somewhere else with the usual ⌘X, ⌘C and ⌘V shortcuts. In addition to the simple paste that inserts the contents of the clipboard at the location of the cursor, you can paste over the existing data. When you Shift-⌘V, the contents of the clipboard will be mixed in, without altering the length of the file (Unless you are at the end of the file.).

4.10 Special pasting

In addition to the simple copy and pasting commands described above, TwistedWave provides a special pasting command whose behavior can be customized in many ways. This command is available in the menu Edit / Special Paste, or with ⌘Y by default, and can be customized with Edit / Special Paste Options….

Special pasting options

The pop-up button at the top of the special pasting options selects the operation that is performed when the Special Pasting command is invoked.

Insert will just insert the contents of the clipboard at the position of the cursor. This is also what does the default paste, available on ⌘V.
Mix will not make make the document any longer, but will mix the clipboard with the contents of the document starting from the position of the cursor. This behavior is similar to the paste over, available on Shift-⌘V.
Replace will replace the selection with the contents of clipboard. This is similar to Mix, except that only the selected area will be affected. The clipboard will be truncated to match the length of the selection. If the selection is longer than the clipboard, the extra sound in the selected area remains unaffected.

What makes the special paste special is its ability to automatically fade in and out the contents of the clipboard, and the document in which it is pasted.

When copying and pasting, by default, TwistedWave will make sure that the edges of the clipboard and the document at the insertion point will fall on a sample value of zero. In most cases, this will prevent a click when pasting.

However, the transition to the clipboard can still seem too brutal, and the automatic fading in and out can help a lot making things smoother. With the options to fade or not the document and the clipboard, a large number of effects can be achieved. In order to immediately see how the fades are applied, a preview representing the document and the content being pasted is shown at the bottom of this options panel. The blue part represents the document, the red part is the contents of the clipboard, and the slopes show where the fades take place.

4.11 Crossfading

Deleting a region from the middle of a file will produce a noticeable cut in the resulting file. This is often not desirable, and a common solution to such a problem is to apply a crossfade.

In order to apply a crossfade in TwistedWave, simply select the region you want to delete, and select Edit/Delete and Crossfade... from the menu, or use the ⌘-Backspace shortcut. A window will pop-up asking you for the duration of the crossfade, and will then proceed to mix the audio before and after the region deleted in order to make the transition smoother.

In addition to the crossfade duration, a pop-up button allows you to use a fade out followed by a fade in instead of the crossfade.

A third option, Crossfade Centered, differs from the Crossfade by the way TwistedWave applies the crossfade.

With the first Crossfade option, the selected area is completely removed, and a piece of sound corresponding to the crossfade duration is taken before the selection, another piece is taken after the selection, and these two are then mixed together to apply the crossfade.

With the Crossfade Centered option, the two pieces of sound are not entirely outside the selection, but centered on the beginning and end of the selection. The benefit is that if you select 10 seconds of sound, for instance, and apply such a crossfade, the total length of the file will be reduced by 10 seconds, and the reduction does not depend on the duration of the fade. This can be very important when editing music, for instance. If you select an area between two beats, you can apply a crossfade without breaking the rhythm.

After a crossfade is applied, you can still manipulate it, change its duration, in and out points. See section 5.2.

Applying a crossfade

4.12 The clip list

When working an a long sound file, it is often necessary to take a few bits out of it, and use them later on to build a new document. This can be done by copying each part of the file you are interested in, and pasting it in a new document, but this can be tedious, and you will end with many document windows obscuring your screen.

The clip list

As a solution to this problem, TwistedWave offers a clip list. You open it from the menu View / Show Clip List . Now, when you zip through the audio and have selected an interesting bit, instead of copying it to the clipboard, create a new clip from it. Simply press the New Clip… button, use the corresponding menu item, or press its key shortcut, simply N. You are then asked to enter a name for the new clip. Once you have filled the clip list, you can play, paste or delete them from the list with the corresponding buttons that appear in front of their names.

Two options can affect the way clips are pasted:

When the Paste over option is checked, the clip is not inserted in the audio file, but mixed with the already existing data, without altering the length of the file.
When the Multiple paste option is checked, when you press the paste button, you will be asked how many times the clip should be pasted. This can be useful when you want to paste a clip many times without having to repeat clicking on the paste button.

5 Editing cuts and crossfades

In order to cut or to apply a crossfade to a part of the sound file, you first have to carefully adjust the selection, and then cut or crossfade (see section 4.11 for applying crossfades). If the result doesn’t sound good, you can undo, tweak the selection a bit, and try again.

TwistedWave offers a better alternative. You can manipulate the cut or crossfades after it has been applied, and preview it without having to undo or redo anything.

Note that cuts and crossfades can only be edited when they are the last operation performed on a sound file. If you perform another operation, the changes you would have made to the cut or crossfade are committed, and you cannot edit them without having to undo and redo them, in the old way.

5.1 Editing cuts

The capture below shows the waveform just after having deleted a part of the audio. A gray vertical bar shows where the cut has taken place, and two arrows are visible on each side of the cut. (Also note that a marker was automatically added by TwistedWave. See section 11.5 to know how to do this.)

These two arrows are in fact two handles, and you can grab and drag them to move the audio in or out of the cut.

Editing a cut

5.2 Editing crossfades

The capture below shows the waveform just after having applied a crossfade. Just as when editing cuts, two arrows appear, which you can grab and drag to adjust the position where the crossfade starts and ends.

You can also grab one of the two vertical bars, and drag it to adjust the duration of the crossfade. Note that depending on the kind of crossfade you have selected, adjusting the crossfade duration can affect the total length of the file or not. See section 4.11 for more information.

Editing a crossfade

6 Playing

When editing an audio file, in order to find out what parts of the file to select, cut, or apply effects to, it is important to know to what parts of the waveform correspond what sounds. When you play the file once, you get a rough idea of what sound corresponds to a given shape in the waveform.

While TwistedWave is playing, it is possible to click in the waveform to move the playhead, or select a specific area you are interested in.

You may find that simply playing a selection is no enough to make accurate selections, and this is where scrubbing comes to the rescue.

6.1 Scrubbing

Scrubbing gives you more control on the position of the playhead, and allows you to listen to very specific parts of the audio and get a clear idea of where specific sounds are located in the waveform.

Scrubbing is performed by clicking and dragging the mouse in the rulers, above the main wave and the overview windows. Depending on your needs, three scrubbing modes can be selected from the preferences window.

Repeat. In this mode, a small area around the mouse pointer is repeated continuously. The audio is played at the nominal speed, and sounds natural.
Velocity. This mode allows you to control the speed at which the playhead moves in the waveform. This mode allows you to play the audio slower or faster than the nominal speed, but this also affects the pitch of the sound. This mode can be very effective when detecting clicks in the audio.
Position. This mode is similar to the velocity mode, except that you have a direct control on the playhead position rather than its velocity. This makes it easier to play a very specific area.

7 Applying effects

This section describes a few effects provided by TwistedWave.

The most interesting effects, however, come from plugins in the Audio Unit or VST format. Using plugins in TwistedWave is described in detail in section 8.

7.1 Amplify

The simplest effect you may want to apply on a signal is to adjust its volume. The amplify effect allows you to choose a gain in decibel to apply to the selected area of the sound file.

7.2 Normalize

When you record from a microphone, or import a sound sample in another way, it can happen that the sound level was not properly adjusted, and appears very low when played, compared to other sound files. You can use the amplify effect to increase the volume by the desired amount. The normalize effect, however, does this automatically. It scans through the entire region, and finds the area where the sound level is the highest. It can then automatically determine by what amount it is possible to amplify without introducing any clipping.

The "Normalize to" pop-up button indicates what level is being normalized. It can be set to one of:

Peak Value. With this setting, TwistedWave look at the sample with the highest peak value, and the sound volume is then adjusted to have the level of this sample match the value specified in the normalization level. With this setting, and a normalization level of 0 dB, for instance, TwistedWave will amplify the sound as much as it is possible without generating any clipping.
True Peak Value. This setting is similar to the above, with the difference that the signal is oversampled in order to detect peaks that happen between two samples. This can be useful to make sure the level does not exceed a certain value after the audio has been converted from digital to analog.
RMS Value. This setting is different from the two previous in that TwistedWave does not look for the loudest part in the audio, but computes the average energy content of the whole audio.
LUFS (BS-1770, R128). This setting is similar to the previous one. The only difference is in the way the average energy content is measured. TwistedWave uses the recommendation ITU-R BS.1770 to compute a value more closely related to the perceived loudness, by attenuating the low and high frequencies before measuring the average energy content.

The next two settings, the maximum gain and maximum attenuation, indicate the maximum amount by which TwistedWave can amplify or attenuate the sound volume to match the given normalization level. If these options are not checked, TwistedWave is allowed to adjust the gain as much as is necessary.

The last setting, maximum peak value is useful when we are normalizing to a target RMS or LUFS value. In these cases, especially if there is a lot of dynamic range in the audio, it is possible for the normalized sound to exceed the full scale, and clip. When this option is checked, TwistedWave will make sure the maximum peak value does not exceed the specified value, even if that means the loudness does not reach the target value.

7.3 Fade in, fade out

These effects will continually adjust the sound volume of the selected area. When applying a fade in, the sound volume gradually increases to its original level. A fade out does the opposite, and decreases the volume to zero.

To see how you can apply crossfades, see section 4.11.

7.4 Custom fade

The custom fade is more customizable than a simple fade in or fade out. Instead of gradually fading from 0 to 100%, the attenuation can be specified at the beginning and the end of the selection.

7.5 Loop crossfade

The "Loop Crossfade" is a special kind of fading where the end of the sample is crossfaded with the beginning. The resulting sample can be played seamlessly in loop.

7.6 Manipulating channels

All the channel manipulation functions are grouped in the Edit menu. A few basic functions allow you to add or remove channels, and convert the document to mono or stereo.

When working on complex documents with more than two channels, it may be hard to get anything useful from these basic functions. For all the complex tasks you will find a tool to reorder the channels in the menu Edit / Reorder Channels…

The channel reordering tool

With the channel reordering tool, you can:

Reorder the channels by drag and dropping them.
Remove selected channels by pressing the delete key.
Add channels by duplicating already existing ones.

7.7 Sampling rate conversion

The sampling rate is the frequency at which the sound is sampled. On a CD, for instance, the sampling rate is 44 100 Hz. It can be necessary to convert the sampling rate if you have a recording at 48 kHz, and you want to record it on a CD, for example.

With a lower sampling rate, the sound data will take less space on the disk, but at the cost of loosing the high frequencies. The highest frequency that can be represented at a given sampling rate is half the value of the sampling rate. On a CD, for example, no sound higher than 22 050 Hz can be represented.

The sampling rate converter

The window that asks you for a new sampling rate offers the option to resample. When it is checked, as it is by default, the wave will be resampled so that it does not sound different (Unless you are converting to a lower sampling rate, in which case you will lose the higher frequencies.). If you uncheck it, TwistedWave will keep the samples untouched, and merely tag the audio as having the new sampling rate. This will result in the sound to be played faster or slower. The file will also sound higher or lower, just as if you were adjusting the speed of a turntable. If you want to change the speed of a file without modifying the pitch, you should use the Change pitch and speed function, described in the next section.

7.8 Change pitch and speed

This function allows you to change the pitch and speed of a sound file. It can be used to change the speed without affecting the pitch, changing the pitch and keeping the original speed, or any combination of pitch shifting and time stretching.

The pitch and speed adjustment interface

By checking the Lock speed and pitch option, you will get the effect you would have when playing a tape or record at a different speed, where both the speed and pitch are affected at the same time.

When the Use ZTX processing engine option is checked, instead of using Apple’s time stretching implementation, TwistedWave uses the ZTX Time Stretch/Pitch Shift technology licensed from Zynaptiq GmbH, http://www.zynaptiq.com/ztx/, (c) Zynaptiq GmbH.

ZTX’ pitch shifting is one of the best pitch shifting implementations, and is an excellent replacement for Apple’s one.

In addition to shifting the pitch of the audio, the ZTX processing engine can also shift the formants. By shifting them with the opposite value of the pitch shift, they will be unaffected by the pitch transformation, and the result will sound more natural.

By checking the Natural pitch option, the formant shift will automatically be adjusted to avoid them being pitch shifted.

7.8.1 Time/frequency localization and the ‘lambda’ parameter

ZTX uses a novel algorithm that can be scaled to provide good time domain localization or good frequency localization, or both. High time localization means that ZTX produces results similar to the time domain pitch-synchronized overlap-add (PSOLA) methods, high frequency localization produces results that are closer to what you get from an improved phase vocoder.

This ability is controlled by the “lambda” parameter. As a rule of thumb, a low Lambda value provides good time localization (good for voice and single instrument recordings), while a high lambda value is good for entire mixes. High lambda values take slightly more time to process but are not considerably slower.

The following lambda settings are available:

Preview. This automatically selects the best time/frequency trade-off for realtime/preview performance. It is the fastest setting but might not provide the best results in all cases.
Voice. Selects full time localization. Good setting for single instruments and voice.
Duet. Time/frequency localization with emphasis on time localization. If a setting of Voice produces echoes this might be a better choice.
Standard. This sets the time/frequency localization halfway between time and frequency domains. It is the best setting for all general purpose signals and should be set as default for non-realtime (non-preview) processing.
Smooth. Higher frequency localization and less time localization. Might be a better choice for classical music than the lower-lambda settings
Classics. Highest frequency localization. This might not be an ideal choice if you’re dealing with signals that have very sharp attack transients but it might be useful for sensitive material such as classics
Transcribe. Special mode for very large stretch ratios (2x to 4x).

7.9 Pitch correction

The pitch correction uses the ZTX Time Stretch/Pitch Shift technology to automatically adjust the pitch of the selected audio. It works by detecting the pitch of each note, and adjusting them to the closest note.

The pitch correction interface

The pitch correction can be altered with the following parameters:

Basic tuning. This value defines the reference tuning for the pitch correction.
Slur speed. Defines the time it takes for the correction to reach the full correction amount. Typically, notes are a bit unstable at the beginning, because the attack phase of a sound has a higher amount of noise, and because singers gradually adjust their tuning after the onset of the note. The slur time makes the pitch correction sound natural because it models this effect. Higher values will yield a slower adaptation time and it will take longer for the correction to produce the corrected pitch.However, longer slur times will also preserve vibrato better.
Quality. This setting allows you to choose a compromise between the sound quality and the processing speed.

7.10 Noise reduction

It is possible to use TwistedWave to remove background noises from a recording.

In order to do that, TwistedWave first has to learn about the noise you want to remove. Select a part of your file that contains only the background noise, and none of the sounds you want to keep, and select Effects / Learn Noise Profile from the main menu.

Once TwistedWave has learned the noise profile, you can select the whole file, or just the part you want to remove noises from, and select Effects / Denoise... from the menu. The Denoise Amount slider allows you to adjust the amount of noise to be removed from the recording. You can click the Play button to preview the effect, and click Apply when you are satisfied with the result.

I suggest you use the smallest amount of reduction that will remove the unwanted noises, in order to avoid removing too much of the sound you want to keep.

8 Audio Unit and VST plugins

Audio Units is Apple’s audio plugin technology. These plugins produce sound effects that can be applied to the selected parts of your documents. A few Audio Units are provided by Apple and are available on every Mac, such as a few filters, equalizer, pitch shifter, reverb and more. Many Audio Units provided by third parties are also available. A comprehensive list is available on macmusic.org.

VST (Virtual Studio Technology) is another plugin technology, created by Steinberg. Similarly to Audio Units, VST plugins can provide many effects, and a comprehensive list is also available on macmusic.org.

8.1 Using a single Audio Unit or VST plugin

An Audio Unit window

Loading a plugin is as simple as choosing it from the Effects / Audio Units & VST menu. TwistedWave will open a window showing the plugin interface.

When you opened an Audio Unit or a VST plugin, it remains active on the current document. That means that when you start the playback, the plugin will be applied to the file in real time. You can even have more than one open at the same time, and they will all process the sound when the document is being played.

In some cases, you want an Audio Unit or VST plugin to be applied not just to one document, but to all the documents open in TwistedWave. This can be done with the Global option. When it is checked, the plugin window will remain visible, and the effect will be applied when playing back any document. This can be useful in particular with visualization plugins. A VU meter, for instance, could be useful for all the documents you could be working on.

If you want to permanently alter the sound file by processing it with the the Audio Unit or VST plugin, you can press the Apply button, and the plugin will process the selected area of the document, or the whole document if nothing was selected.

Note that when the file is saved, the active plugins will not be saved, and won’t affect the file being saved. If you want to save the result of processing the document with a plugin, you should apply it first.

In the top right hand corner of the plugin window, a pop-up button offers you to choose what to do with the tail of the effect. The tail of the effect is the sound produced by the Audio Unit or the VST plugin after it has finished processing the selection. If you use a delay to add an echo, for instance, it might take a few seconds for the echo to repeat until it has completely disappeared. There are three ways in which TwistedWave can deal with this tail:

Truncate tail. With this option, the plugin will stop processing once it has reached the end of the selection, and the tail will be discarded.
Insert tail. The tail will be inserted in the document, at the end of the selection.
Merge tail. The tail will be mixed with the contents of the document after the selection. Note that this option will alter a part of the sound that is not selected by mixing the tail of the effect to it.

8.2 Saving presets

You can press the Save button at the top of the plugin window to save the current preset. You will then be asked to name the new preset. You can access saved presets from the Load Preset pop-up button.

Presets are saved as files in a folder corresponding to the current plugin. Selecting Manage Presets… from the menu will open this folder in the Finder. This allows you to rename or delete presets. You can also put them in sub-folders, and they will appear in sub-menus when you open the presets menu.

8.3 Using the effect stack

With the effect stacks, TwistedWave provides a convenient way to group plugin together. This makes it easy to save their presets as a whole, reorder the plugins, or apply all of them at once.

Select New Effect Stack… from the Effects menu, and TwistedWave will open an initially empty window. The Add Effect menu allows you to choose the Audio Unit or VST plugins you want to add to the stack. When you have added a few effects, you can start previewing them by pressing the Play button, and all the effects in the stack will be applied in real time on the selected area of the document. That means that you can change the parameters of all the Audio Unit and VST plugins before committing the changes to the document.

$The Effect Stack with two effects being\n edited$

The Effect Stack with two effects being edited

The effect stack works as a single plugin window (see section 8.1) in the way that an effect stack window is always active on the current document. The effects it contains are always applied when playing back the document.

In the same way that you can save Presets for individual Audio Unit or VST plugins, you can save a whole effect stack. In addition to the Load Stack pop-up button, you can load an effect stack directly from the main Effects menu. A Load Stack menu appears below the New Effect Stack… when some saved stacks are available.

9 Generating Sounds

9.1 Signal Generator

The signal generator is a tool that allows you to generate simple signal patterns. The options allow you to select the shape of the wave (sine, rectangle, triangle or sawtooth), its frequency and amplitude.

If you have checked the Sweep option, you will have the ability to generate chirps by selecting an end frequency. The frequency of the generated signal will change progressively from the original frequency parameter to the end frequency you have specified.

When the Automatically adjust frequency to avoid clicks option is enabled, TwistedWave will make sure the end of the generated signal corresponds exactly with the end of one period. This can be useful to avoid clicks at the end of the signal. The frequency of the signal, or the end frequency in the case of a sweep, will be adjusted to make sure the wave reaches zero at the end of the generated audio.

In addition to periodic signals, the signal generator can produce white or pink noise. These are two kinds of noises where the energy is distributed across all frequencies in the spectrum.

When you have adjusted the signal parameters, clicking Apply will add the generated signal to the current document in one of three ways, as chosen from the Action pop-up button.

Insert will insert the generated signal at the position of the cursor.
Replace Selection will replace the contents of the selection with the generated signal.
Mix with Selection will keep the contents of the selection, and mix it with the generated signal.

Signal Generator

9.2 Speech Synthesis

The speech synthesis effect uses Apple’s built-in speech synthesis to convert a text to speech.

Enter some text, and press OK to insert the spoken text in the current document at the position of the cursor.

Speech Synthesis

A number of different voices are available to synthesize the speech. They can also be further refined by adjusting the speech speed and the pitch of the voice.

It is possible to install more voices from the System Preferences. In the accessibility preferences, click "Speech", and then in the "System Voice" pop-up button, select "Customize...".

10 Recording

In addition to being able to import samples in many different formats, TwistedWave allows you to create new samples by recording from any of the audio devices connected to your Mac, such as a microphone.

You can start recording by selecting Audio / Start Recording from the menu, or by pressing the Record button from the toolbar. Note that this function is available only when TwistedWave is ready, and is not currently busy reading a file, applying a filter, or performing any other background task.

When the recording session has started, the audio is inserted in the current document at the position of the insertion point, and the level meters show the peak levels of the audio being recorded. This is intended to help you adjust the input level to a good value, and making sure no clipping takes place.

10.1 Punch-in recording

Another recording mode is available by selecting Audio / Punch-In Recording. When that option is selected, TwistedWave will play a couple of seconds of audio before the cursor position, and when it reaches the cursor, it stops playing audio back and switches to recording. This can be very useful to re-record a section of a longer recording.

When doing a punch-in recording, by default, the recorded audio replaces the existing audio. It is possible to change that behavior, and have the new audio be inserted instead. This can be configured from the preferences window, in the Recording section of the Editing tab.

If you are initiating a punch-in recording while the selection is active, the recorded audio will never be inserted, but will instead replace the selected region. Once the recorder has reached the end of the selection, it will automatically stop recording.

This can be useful when you want to re-record a single word or sentence, for instance, and replace it with a new recording of the same duration.

10.2 Glitch detection

When TwistedWave is recording, the audio is provided by macOS one buffer at a time. In some rare cases, when using very small buffers and / or when your Mac is heavily loaded by other CPU intensive applications, it is possible that TwistedWave misses a buffer, and a glitch is introduced in the audio.

TwistedWave has the ability to detect these glitches and will tell you how many glitches have occurred at the end of the recording session.

In order to help you identify the glitches, TwistedWave will also automatically place a marker where they happen, unless you are using TwistedWave Lite where support for markers is not available.

If you are experiencing glitches in your recordings, a good way to avoid them is to increase the buffer size, in the audio devices preferences (see section 18.3).

11 Markers

Markers are a kind of index that can be used to identify a specific place in the audio file. They can be given a name, which makes it obvious what they point to.

Another use for markers is to indicate the beginning of each track which might compose a sound file. When preparing a CD, the application responsible for burning the CD might expect a list of markers to indicate the beginning of each track.

11.1 Manipulating markers

The capture here shows a document with a few markers. The markers are represented as a vertical line in the waveform. Additionally, an arrow and a label are shown in the area above the time ruler. Within this area, you can:

Add a marker by clicking anywhere in the ruler.
Move a marker by dragging it within the ruler.
Remove a marker by drag and dropping it out of the ruler.
Manually enter its position and label by double clicking on a marker.

Markers

A convenient way to create markers is to click Markers / Add Marker from the menu, or press the shortcut key M. This command will automatically add a marker at the current cursor position.

Because this also works when TwistedWave is playing or recording, you can play a whole file, and mark places you want to inspect and edit at a later time.

TwistedWave can also automatically add markers when certain operations are performed. See section 11.5.

11.2 Importing/exporting markers

When an audio file is burnt to a CD, markers are often used to mark the beginning of each track. Although TwistedWave can save markers in WAV or AIFF files, some software prefer having the markers saved in a separate file. From the Markers menu, the options Import Markers… and Export Markers… allow you to manage CUE files containing markers. The CUE files are commonly recognized by CD burning applications, and contain the necessary information to split the audio file in several tracks.

11.3 Markers window

From the menu, the option Markers / Markers Window opens a window showing a list of all the markers present in the current document.

By double clicking on the items in this list, you can easily rename the markers, and set their positions to a precise time.

The markers window

11.4 Splitting by markers

Because markers are often used to indicate the beginning of individual track in an audio file, an area between a pair of markers corresponds to a track, and in order to work on individual tracks, it can be useful to split the tracks, and save them in individual files. This can be done by selecting Markers / Split by Markers from the menu.

By default, the names of the files generated correspond to the names of the markers, prepended by the number of the track. You can also manually edit these names, or uncheck them if you don’t want to save a specific track.

Splitting by markers

11.5 Automatic markers

Some operations, such as cutting, or applying a crossfade can have a dramatic effect on the sound file. Therefore, it can be important to always know where such operations were applied. A convenient way to always remember a place is to insert a marker.

When the option Markers / Auto Mark Cuts is enabled, TwistedWave will automatically add a marker whenever you cut a part of the file, or apply a crossfade (Delete or Delete Crossfade… from the Edit menu).

12 Speech recognition

12.1 Recognizing speech

When working with spoken word, given enough experience, it is possible to identify sentences and even individual words from the shape of the waveform, but you have to listen to it in order to know what word corresponds to each part of the waveform.

By using speech recognition, TwistedWave can automatically show the content of your audio in plain text, enabling you to see the contents of your recording without having to play it back.

There are two ways to trigger a speech recognition. You can start it manually on an existing audio file or recording (or a selected area thereof) by selecting Speech / Recognize Speech from the menu.

You can also have TwistedWave recognize the speech in real-time while audio is being recorded by selecting Speech / Record With Speech Recognition from the menu.

Speech recognized

12.2 Configuring Speech Recognition

Before you can use it, you need to configure the language or model to use for the speech recognition.

TwistedWave can use one of two speech recognition engines. As described below, the Whisper based speech recognition is currently only available for the manual speech recognition. For that reason, it is possible to select two different models, one for the real-time speech recognition, and a different one for the manual speech recognition.

12.2.1 Apple’s Speech Recognition

Apple’s speech recognition, available since macOS 10.15, can be used to perform the speech recognition of your audio files and recordings directly on your Mac, without requiring an internet connection, or sending the audio to Apple’s servers.

Before you can use Apple’s speech recognition, you need to select the language that has to be recognized, either from the speech tab of TwistedWave’s preferences, or by selecting Speech / change Recognition Model... from the main menu.

For a language to be available for selection in TwistedWave, it has to be configured in macOS’ system preferences by following these steps:

Open System Preferences.
Click Accessibility.
In the sidebar, scroll down and choose Voice Control.
Tick the Enable Voice Control checkbox.
You will be prompted by macOS to enter your password.
macOS may need to do a one-time download. Wait for it.
You may untick the Enable Voice Control checkbox.

12.2.2 OpenAI’s Whisper Speech Recognition

Since version 29.3, TwistedWave also offers the possibility to use OpenAI’s Whisper speech recognition engine.

Whisper also runs locally on your Mac. Its main advantage is that depending on the model you choose, it can be either much faster, or much more accurate than Apple’s speech recognition.

Instead of a language, in order to use Whisper, you need to select one of different speech recognition models with different speed / accuracy trade offs. Some models are English only, and will only recognize speech in the English language, presumably with better accuracy. The other models are multilingual, and will automatically recognize the language in the speech.

The only limitation of Whisper is that it is currently not possible to use it for real-time speech recognition, and you can only use it manually on an existing audio file or recording (or a selected area thereof) by selecting Speech / Recognize Speech from the menu.

12.3 Speech markers

When TwistedWave recognizes speech, it shows the text content of each word that was recognized in the file. These are stored internally as markers, so they behave exactly like markers. If you copy and paste audio around, the speech markers will be copied along with it.

The speech markers are in fact regular markers behind the scenes. TwistedWave can tell a regular marker from a speech marker by placing a double quote in front of the marker name. If the word hello has been recognized, for instance, a marker named "hello will be stored in the audio file.

Because you may want to keep the speech recognition to continue working with it at a later time, TwistedWave also saves these markers along with the audio, exactly as it would save regular markers.

When you send your audio files to someone else, if they don’t open it with TwistedWave, they may be annoyed by the number of markers placed by TwistedWave’s speech recognition. For that reason, it is possible to remove the speech markers by selecting Speech / Clear Speech Recognition from the menu. Alternatively, you can prevent TwistedWave from saving the speech markers by disabling the option to Save Speech Recognition With Files either from the Speech menu, or from the Speech tab of the preferences window.

12.4 Synchronizing with a text script

Text Script window synchronized with the waveform

Once the speech in your audio file has been recognized, and the speech markers are visible in the main waveform, it is possible to synchronize the audio file with a text script, either in the .txt, .rtf, .rtfd or .pdf format, then TwistedWave can open it, and show it along with your audio file.

Select Speech > Load Text Script... from the menu, or simply drag and drop the text script over the editor window, and TwistedWave will load the script, and match the text it contains with the speech that was recognized in your recording.

Once the script was loaded, it will highlight in yellow the words that have been recognized in the recording.

12.4.1 Synchronizing cursor and selection

Clicking one of the recognized yellow words in the text script will automatically place the cursor in the audio recording at the corresponding position. Likewise, placing the cursor in the audio recording will scroll the script and select the corresponding word in the text.

Similarly, selecting a region in either the text script or the audio recording will select the corresponding region in the other window.

12.4.2 Detecting multiple takes

When recording a long text, it sometimes happen that a word is misread, and you need to say the whole sentence again.

When that happens, you have to edit the audio file, and delete the incorrect sentence. Manually placing a marker when you record and notice the mistake will make it convenient to highlight the places that need editing.

Because TwistedWave knows the script you are reading, it can automatically detect the instances where you go back and say a sentence a second time. After you are done recording, it will automatically place markers named "jump" to indicate the places where you have jumped to another part of the script.

12.5 Speech recognition while recording

Instead of manually recognizing the speech after it has been recorded, it is possible recognize in real time while the audio is being recorded.

You can force TwistedWave to always recognize speech in the recorded audio by enabling the option to Recognize Speech When Recording, but if you already have a text script loaded, as described above, TwistedWave will always recognize the speech in the recorded audio.

The recognized speech will be matched with the script as you speak, and the text will scroll automatically, allowing you to read long texts without having to manually scroll it.

13 Metadata

Many audio file formats can store metadata in addition to the sound data itself. Metadata can consist in several pieces of information you would like to record with your document. In the case of a song for instance, you may want to record the title of the song, the name of the album and artist, and so on.

You can edit this metadata with the metadata edit panel, which can be open from the menu File / Show Info…, or by clicking the Info button from the toolbar.

TwistedWave can read and save music metadata from files in the aiff, wav, mp3, mp4, flac and ogg/vobis format.

3 tabs of the metadata edit panel

13.1 BWAV and Soundminer

In addition to music metadata, the edit panel can show up to two additional tabs for editing BWAV and Soundminer metadata. These metadata are not specific to music, and can be used to store more general information, such as a description of the sound, what kind of microphone was used for the recording, where the sound was recorded…

Note that BWAV metadata can only be saved in .wav files, and this tab will only be present when editing a file in this format. Soundminer metadata can be present in .wav and .aiff files.

13.2 FLAC metadata

Although FLAC does not have native support for BWAV and Soundminer metadata, TwistedWave still allows you to edit and save FLAC files with this kind of metadata.

Since version 1.2.1, the FLAC encoder has support for metadata found in the WAV and AIFF files. When used with the option --keep-foreign-metadata, the encoder will keep the metadata from the original file in the FLAC file. TwistedWave takes advantage of this possibility, and will save the BWAV and Soundminer metadata chunks in the FLAC files. When the FLAC files generated by TwistedWave are decoded with the command line FLAC decoder, and if the --keep-foreign-metadata option is used, the BWAV and Soundminer metadata chunks will be restored in the generated WAV file.

14 Exporting the waveform

In addition to exporting a sound file in a different audio format, TwistedWave offers the ability to export its waveform as a PNG image. Exporting the waveform image with TwistedWave is not very different than taking a screenshot. Where this can be very useful, however, is when this is done in conjunction with the batch processing. This is a very easy way to generate a small waveform image for many files in a single step.

You can read more on batch processing in section 16.

Exporting the waveform

The waveform export window allows you to customize the color scheme for the exported waveform, or to select an existing one from the same presets that are used to customize the appearance of the waveform in the document windows.

The rendered waveform looks the same as it does in the regular editing windows, unless you add some anti aliasing. The image here shows the effect of the anti aliasing on the waveform rendering.

$The effect of antialiasing on\n the exported waveform$

The effect of antialiasing on the exported waveform

15 Video sync

Even though TwistedWave is an audio editor, it has always been able to import the audio track from a movie, and edit it just as any regular audio file.

With the "Video Sync" feature, support for video in TwistedWave is going up a step, and it is now possible to do much more than just loading an audio track, as described in the following subsections.

15.1 Save the audio track back to the original movie file

When you open a movie file in TwistedWave, it will show the audio editor with the sound track from the movie. It will also open a second window showing the video track.

The video will play when you play the audio, and will jump to the corresponding frame when you place the cursor anywhere in the audio waveform.

When you want to save your audio file, in addition to the usual audio formats, you can select a movie format. When you save a movie, TwistedWave will keep the original video, add the edited sound track, and save it back to the original or to a new movie file.

If the option Video > Video Edits Follow Audio Edits is not enabled (see next section), then the video is not re-encoded, and the full image quality is preserved, allowing you to save the same video multiple times without affecting the image quality. You should read section 15.4 for more on audio and image degradation when saving files.

15.2 Automatically edit a movie by matching the audio edits

If you select the option Video > Video Edits Follow Audio Edits from the main menu, or by clicking the video edit icon in the toolbar of the video window, then the video will not be saved untouched to the destination file, but will instead follow all the edits that have been done to the audio, such as:

Cutting the audio will cut the corresponding video,
Copy / pasting audio around will also copy the images,
Reversing the audio will reverse the video,
Time stretching the audio will slow down or accelerate the video.

You should keep in mind that because the video is being edited, it has to be re-encoded by TwistedWave, and it may result in a degraded image quality. Every time you close, reopen and edit the movie, such degradation will occur. TwistedWave is smart, though, and is usually able to re-encode only the parts where edits did occur, and re-use the encoded images from the original file, as you can see in section 15.4.

15.3 Create a movie by linking an audio file to an existing video

There are cases where you have an existing audio file that you want to use as the sound track for a movie, or you want to record in an empty file, and have the recorded audio saved with a video. Select Video > Link Video... from the main menu, and you will be able to select the video file you want to use with your sound track. The options to select the movie file formats in the file save dialog will then be available.

15.4 Re-encoding and image degradation

When you open a compressed audio file, such as an mp3 file, edit it, and then save it again, the audio has to be re-encoded and this can result in some degradation of the original audio. For that reason, it is advised to use a lossless file format when you need to re-open and edit the same file multiple files.

Similarly, it is possible to store the audio track uncompressed in a movie file. The MPEG4 file format doesn’t support this, but it is possible to store the audio uncompressed in QuickTime (.mov) and Matroska (.mkv) files. You should use one of these and select no audio compression instead of AAC if you need to edit the same file multiple times.

The same happens with videos, and re-encoding of video frames should be avoided as much as possible. That being said, TwistedWave can avoid re-encoding video frames in many cases, such as:

When the video is not edited (the Video > Video Edits Follow Audio Edits option is disabled, see section 15.2), and the destination file format uses the same video codec as in the original video, then TwistedWave can save the video without re-encoding anything.
When the video is edited and the destination file format uses the same video codec as in the original video, then TwistedWave needs to re-encode the video, but only the parts of the video that are edited need to be re-encoded. If you do a single cut in a video file, for instance, then only a small region around the cut has to be re-encoded, depending on the position of key frames in the video file.
When the video is edited and the destination file format uses the same video codec as in the original video, there are some cases where TwistedWave is able to re-use all the original frames and not re-encode anything, preserving the full image quality. This is possible when all the images in the video track are key frames. This can happen when the original video was encoded with a codec that only uses key frames, such as "Motion JPEG", or with the h264 codec configured to use only key frames (some digital cameras provide that option with the ALL-I or ALL-Intra setting).

16 Batch processing

TwistedWave’s batch processing abilities make it very easy to apply a number of effects or other transformations to a large number of files.

The batch processing window, shown in the capture here, is composed of two parts. The batch script is built in the upper half, and the set of files on which to apply the script is selected in the lower half.

Batch processing window

16.1 Defining the batch script

The upper part of the batch processing window shows mainly a list of all the actions that have to be performed on the selected files. When the batch process is run, the actions are run one by one, from the top of the list to the end on each of the selected files.

This list is initially empty, and a new action is added by selecting it from the Add Action pop-up button in the lower left hand corner. Most of them are the same effects that can be found in the main menu when working on individual files.

In order for the batch script to do anything useful with the files, it has to contain at least one of Save, Export Waveform Image, or Analyze. You can even have more than one Save, if you want to convert a bunch of files to different file formats at the same time, for instance.

The behavior of most of the actions that can compose a batch script can be further customized by clicking the Edit button sitting next to their name. Than allows you to indicate the amount by which to change the pitch or speed of a file, for instance.

In the lower right hand corner, another pop-up button allows you to save batch scripts, or load a previously saved one.

16.2 Saving processed files

In order to save the files resulting from the batch process, you have to add at least one Save action at the end of the process. As shown in the capture, the Save action offers a number of options to select the file format, destination and name.

Saving options

It offers the possibility to add a prefix or a suffix in order to alter the file name.

The location of the saved file, by default, is the same as the original file. It is possible, however, to select another folder.

The format of the saved file can be either the original file format, or another format which you can select from the file format pop-up button.

Note that if you select both the original location and format, the processed files will replace the original files, and such an action should be performed only after the batch process was tested, and verified to work as expected. If you select a different location or file format, any already existing file will only be overwritten if the Allow overwrites option was enabled.

16.3 Running the batch

When a batch process has been defined, you can add a few files to the list by clicking the Add Files... button, or more simply by drag and dropping them directly to the list.

You can add single files as well as folders to the list. When you add a folder, all the sound files it contains are looked for, and the number of file it contains is indicated in the list.

Also note that when a folder has been added to the list, when a Save operation is being applied, if the destination is not the original location, the original directory tree is reconstructed in the destination folder, with the processed versions of the selected files.

You can then start the batch by clicking the Start Processing button. If you want to test the batch, and run it on a single file without saving anything, you can click the Test button in front of a file name, and the processed file will open in an unsaved TwistedWave document window.

16.4 Concurrent tasks

By default, TwistedWave will process the files one after the other. For a long time, Mac computers have been equipped with more than one CPU cores, and taking advantage of them to process the audio files in parallel can significantly speed up the process.

Next to the Start Processing button, there is a pop-up button that says N Concurrent Tasks. You can click it, and specify the number of tasks you want to use. The number can go up to the number of parallel threads that can run on your CPU.

Some tasks don’t require much CPU, and are rather limited by the speed of your storage. If you want to convert many MP3 files to WAV, for instance, using multiple concurrent tasks is not likely to speed up the process a lot, because the CPU will be waiting for the disk to finish writing the WAV file. This is especially true with slower hard disk storage.

Another thing to take into account is that TwistedWave will use temporary cache files when processing an audio file. If you process many files in parallel, and these files are large, the cache files may exceed your available storage, and TwistedWave will be unable to complete the batch process.

16.5 Splitting files

The result of applying the batch script on a file will result in a processed file which will usually be saved by one or more Save operation at the end of the script.

Two batch processing actions (namely Split by Markers and Split Channels) have the special ability to split an audio file into several parts. When processing a file results in more than one part, the Save operations will not save a single file, but the different parts will be saved in different files.

17 TwistedWave iOS Remote

The TwistedWave Remote

The TwistedWave remote is an application for iPhone or iPod that can be downloaded from the App Store. It will connect to a TwistedWave application running on a mac, and allow you to remotely control it and perform a few simple tasks, such as playing, recording or placing markers.

18 Preferences

The Preferences window enables you to customize the behavior of TwistedWave. The different options available are described in the sections below.

18.1 General settings

The Preferences Window (general settings)

18.1.1 Time display

Some WAV or AIFF files can contain a time code information specifying the time reference of the file. When this information is available, TwistedWave uses it, and the timeline and cursor position displays will use it as the time origin. Even if this information is available, it may be desirable to work with times relative to the beginning of the file, which can be done by unchecking the option Use time origin from audio file.
By default, the values displayed in the timelines and for the cursor position in the toolbar are in minutes, seconds, and fractions of a second. With the "Time format" option, you can also select to see the time information in sample count, or display a SMPTE time code value.

18.1.2 Level Meter

Two settings allow you to customize the loudness indication visible below the level meter. By default, this value shows the RMS level in decibels.

The value displayed corresponds to the energy integrated on a window whose length can be specified with the "integration window" setting. With the default of 10 seconds, the value displayed represents the loudness of the 10 previous seconds of audio.

The "Loudness indication" setting allows you to select between the RMS and LUFS measurement for the loudness indications.

When you select the LUFS, before integrating the energy contained in the audio, a filter is applied to the sound in order to discard the frequencies that are not perceived by the human ear, and don’t contribute to the loudness.

18.1.3 Registration

When you register TwistedWave, either by entering a serial number, or by purchasing directly from within the application, TwistedWave becomes registered only for the user that is currently logged in on your computer.

In the preferences window, an option allows you to make the registration available to all the users.

18.1.4 Check for update

TwistedWave can automatically check for updates. It is recommended to configure automatic updates, in order to get the latest bug fixes, and keep TwistedWave compatible with the latest macOS updates.

By checking the Include beta updates option, in addition to the official updates, you will be offered beta updates that contain the latest bug fixes and improvements.

If you are having an issue or have a suggestion regarding TwistedWave, especially if you are using a beta version, you are encouraged to let me know with the contact information available on TwistedWave’s support page.

18.1.5 Scratch disk location

When you edit a file in TwistedWave, all the audio is stored, uncompressed, in temporary files on your computer’s disk, and will remain there until you close the editor window. If you work with very large files, or need to open many files simultaneously, TwistedWave may need to use a lot of temporary storage.

TwistedWave uses the system disk by default for the temporary storage. If you need more space than is available on the system disk and you have another volume connected to your computer, you can change the scratch disk location to tell TwistedWave to use the disk you want.

18.1.6 TwistedWave iOS Remote

When you have enabled the TwistedWave Remote, you can connected to your running TwistedWave instance with the iOS application, available on the app store.

You can see how to use the remote in section 17.

18.2 Editing preferences

The Preferences Window (editing preferences)

18.2.1 Editing preferences

The "Auto extend to crossing" option allows you to enable or disable the ability of TwistedWave to automatically extend the selection to the next zero crossing points in the sound. This capability is described in more detail in section 4.6.
The "Lock channels together" option can be very useful when working on multichannel files. When this option is checked, selecting an area will always select all the channels at the same time.
By default, when zooming with the scroll wheel, TwistedWave makes sure the insertion point remains in the center of the window so that it is always visible. If you check the option "Zooming centered on mouse", zooming in and out will be centered on the mouse position instead. This allows you to control the point where you are zooming in by moving the mouse while you are zooming. This can be an efficient way to find a specific location in the waveform.

18.2.2 Playing

The "Auto scroll when playing" option allows you to enable or disable the automatic scrolling of the waveform when playing. Disabling it can be useful when you are editing the selection, and don’t want to loose the current position just to start the playback in order to check what you have selected.
When the option "Move cursor back after playing" is checked, the cursor remembers its position, and comes back to its original place after playing.
The "Scrubbing mode" option allows you to select the behavior of the playhead when dragging the mouse to perform scrubbing. The different scrubbing options are described in section 6.1.

18.2.3 Recording

Two options allow you to control the behavior of punch-in recordings. (You can read more on punch-in recording in section 10.1.)

The first option allows you to configure the duration of the pre-roll. This is the number of seconds of audio that are played back before TwistedWave starts recording.

The second option, disabled by default, tells TwistedWave that the recorded audio should be inserted in the file instead of overwriting what is already there.

Note that this option has no effect if the selection is active when you start a punch-in recording. In that case, the recorded audio will replace the selection, and the recorder will stop automatically when it reaches the end of the selection.

18.2.4 Transient threshold

This slider allows you to adjust the threshold above which TwistedWave can detect transients. The higher the value, the more pronounced the transients have to be in order for TwistedWave to be able to detect them. The value you use may depend on the kind of document you are working on. If the transients are very marked, you can use a higher value, in order to avoid detecting unwanted transients. Using the transient detection in TwistedWave is described in more details in section 4.8.

18.3 Audio devices

The Preferences Window (device settings)

18.3.1 Selecting audio devices

At the top of the devices tab of the preferences window, two pop-up buttons allow you to select which input and output devices are used for recording and playback.

If you select System Settings, TwistedWave will follow the audio device used by the system for recording and audio playback.

If you select a specific device from the list, TwistedWave use that one instead. If the device selected for recording is not available, if it is not plugged-in or switched on for instance, TwistedWave will warn you, and ask that you select another device. That will prevent you from recording from an unwanted device, and notice after the fact that you have been recording from your Mac’s built-in microphone instead of your $1000 microphone.

18.3.2 Play through

When recording, you may want to monitor the sound being recorded. By checking the option to Play through when recording, TwistedWave will automatically play the audio fed to the input device when you start recording.

18.3.3 Buffer size

A pop-up button enables you to adjust the buffer size. This instructs TwistedWave about the size of the chunks of audio that should be used when communicating with the sound card when playing or recording. This setting is useful mainly when the Play through option is used. When playing through, a small delay may be noticed between the time the audio is recorded, and when it is played back on the output device. By selecting a smaller buffer size, this delay can be reduced.

18.3.4 Matching the sample rate

When you record audio with TwistedWave, it will automatically adjust your audio interface’s sample rate to match that of the file you are recording, in order to avoid sample rate conversions that could slightly degrade the audio quality.

When playing audio back, the sample rate of the audio interface is not adjusted by default. You may have another audio software using the same audio interface, and fighting with TwistedWave to use another sample rate. In order to avoid that kind of conflicts, you have to manually check the Match the file sample rate when playing checkbox.

18.3.5 Detect glitches

The Detect glitches option enables TwistedWave’s glitch detection when recording. You can read more about glitch detection in section 10.2.

18.3.6 Selecting channels

When TwistedWave records from a device that has more channels than the file you are currently working on, it uses only as many channels as necessary. If you have a 16 channel input device, and record a stereo file, for instance, TwistedWave will record from the first two input channels of your input device. However, you may not always want to use the first two channels. In order to solve this, press the button "Select Channels…", and TwistedWave will open a sheet allowing you to specify the order in which the channels should be used for recording.

This works exactly in the same way for playing, where you select the channels to be used for the output.

18.4 Speech recognition

This tab allows you to configure the speech recognition integrated in TwistedWave. You can read more about the speech recognition in TwistedWave in section 12.

$The Preferences Window (speech recognition\n settings)$

The Preferences Window (speech recognition settings)

18.4.1 Language

Before the speech recognition can be used, it is necessary to select the language of the spoken text.

There are not many languages available by default in the languages menu. If the language you are using is not listed, and because the speech recognition engine used by TwistedWave is the same as that used by macOS’ dictation service, you can make a new language available by selecting it as the language used for dictation. Open macOS’ Keyboard preferences, and in the Dictation tab, you can configure a new language. It may take a few minutes for the new language to be downloaded in the background before it is available in TwistedWave.

You don’t have to restart TwistedWave in order to use the newly available language. Closing and reopening the preferences window is enough.

18.4.2 Speech recognition when recording

By enabling the option to Recognize Speech When Recording, it is possible to have TwistedWave automatically recognize the speech while you are recording, even if a text script was not loaded.

The speech recognition is always done if you are recording with a text script loaded.

18.4.3 Saving speech markers

It is possible to disable the Save Speech Recognition With Files option in order to save audio file without the speech recognition markers. The speech recognition adds many markers in the audio file, and it may be inconvenient when the file is opened in an editor that doesn’t display the speech markers in a special way, like TwistedWave does.

19 Keyboard shortcuts

Most of the keyboard shortcuts available in TwistedWave are attached to menu items, and can be found by browsing the main menu. Additionally, some shortcuts are not attached no any menu item, and are described below.

19.1 Zooming

The + and - keys can be used to zoom the waveform in or out. The same effect can be achieved with the up and down arrows.

The temporary zoom is activated with the Z and Shift-Z shortcuts. The process is explained in detail in section 3.4.

19.2 Manipulating the selection

Clicking and dragging the mouse over the waveform selects an area of the file.
Shift+clicking will edit the selection by dragging the edges of the already existing selection.
Option+clicking will add a region to the already existing selection.
Option+Shift+clicking will remove a region from the already existing selection.

19.3 Moving the cursor

When a selection is present, the left and right arrows will move the cursor to the edges of the selection.

When nothing is selected, the arrows will move the cursor to the beginning or the end of the file.

19.4 Playing

Pressing the space bar will start playing the audio at the position of the cursor. If a region of the file is selected, only that area is played back.

Pressing space again will stop the playback.

20 Feeling at home

When you start working in a new environment, it takes some time getting used to it before you are able to be fully efficient. With a new application, this means getting to know how it works, what it can do for you. If you were used to a similar application, the differences between TwistedWave and what you were used to could complicate the matter. In order to leverage your knowledge of other application, it is possible to change the keyboard shortcuts to ones you are more familiar with. It is also possible to change the colors of the waveform, and make it suit your tastes better.

20.1 Changing colors

In order to change the color of the waveform, you have to open the color editor, located in Edit / Edit Colors…. This will open a window with a list of color swatches and short descriptions. You can then click to edit the colors, and see how it affects the waveform in real time.

There are two columns of color swatches in the editor. The first column is used in light mode, and the second column defines the colors used in dark mode. You can configure macOS to use either the light or dark mode from the system preferences.

Here are a few things to note regarding the colors:

When changing the wave color, you have the option to change both the center and edge color. Having two different colors will fill the wave with a gradient, making it look better.
Similarly for the background color, you have to set the top and bottom color in order to define a gradient.
In order to change the color of the selection, you have to specify the color of the selected wave, and the selected background.

The Color Editor

The color editing window is also the place where you can change the size of the font used in the timelines (the rulers above the main wave and overview windows). You may also like not to have the font antialiased, especially for small sizes.

Similarly to the Audio Unit presets, the Color Schemes pop-up button at the top of the window allows you to save and load color schemes. The color schemes are saved in individual files, and the Manage Color Schemes… option opens a Finder window showing the folder containing all the presets you could have saved. You can remove these files, copy them on another computer or exchange them with friends. These changes will automatically be reflected in the pop-up menu.

20.2 Changing shortcuts

In order to change the shortcuts, select the menu Edit / Edit Keyboard Shortcuts…. The main menu will enter in shortcut edit mode, and a window will invite you to select an item from the menu in order to change its associated keyboard shortcut.

In shortcut edit mode

When you have selected a menu item, a second window shows you the current and default associated shortcuts (if any), and waits for the next key strokes to use as a new shortcut. You are also offered the option to remove the current shortcut, or use the default.

Editing a shortcut

21 Copyright notices

21.1 ZTX Time Scaling technology

ZTX Time Stretch/Pitch Shift technology licensed from Zynaptiq GmbH, http://www.zynaptiq.com/ztx/, (c) Zynaptiq GmbH.

21.2 LAME

TwistedWave uses LAME for encoding mp3 files. LAME is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The LAME source code can be obtained at lame.sourceforge.net

21.3 TwoLame

TwistedWave uses TwoLame for encoding mp3 files. TwoLame is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The TwoLame source code can be obtained at www.twolame.org

21.4 FFMpeg

TwistedWave uses FFMpeg for decoding files in various formats. FFMpeg is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The FFMpeg source code can be obtained at ffmpeg.mplayerhq.hu

21.5 libsox

libsox is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The libsox source code can be obtained at sourceforge.net

21.6 Guile

Guile is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The Guile source code can be obtained at gnu.org

21.7 id3lib

Id3lib is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The id3lib source code can be obtained at id3lib.sourceforge.net

21.8 libsndfile

Libsndfile is licensed under the LGPL (GNU Library General Public License). A copy of the LGPL can be found at gnu.org.

The libsndfile source code can be obtained at www.mega-nerd.com

21.9 Ogg/Vorbis

TwistedWave uses libvorbis and libogg. They are used for reading and writing Ogg/Vorbis files.

Copyright (c) 1994-2004 Xiph.org Foundation

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:

- Redistributions of source code must retain the above copyright
  notice, this list of conditions and the following disclaimer.

- Redistributions in binary form must reproduce the above copyright
  notice, this list of conditions and the following disclaimer in the
  documentation and/or other materials provided with the distribution.

- Neither the name of the Xiph.org Foundation nor the names of its
  contributors may be used to endorse or promote products derived from
  this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

21.10 FLAC

TwistedWave uses libFLAC and libFLAC++. They are used for reading and writing FLAC files.

Copyright (C) 2000,2001,2002,2003,2004  Josh Coalson

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:

- Redistributions of source code must retain the above copyright
  notice, this list of conditions and the following disclaimer.

- Redistributions in binary form must reproduce the above copyright
  notice, this list of conditions and the following disclaimer in the
  documentation and/or other materials provided with the distribution.

- Neither the name of the Xiph.org Foundation nor the names of its
  contributors may be used to endorse or promote products derived from
  this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

21.11 WavPack

TwistedWave uses WavPack. It is used for reading and writing WavPack files.

Copyright (c) 1998 - 2008 Conifer Software
All rights reserved.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:

- Redistributions of source code must retain the above copyright
  notice, this list of conditions and the following disclaimer.

- Redistributions in binary form must reproduce the above copyright
  notice, this list of conditions and the following disclaimer in the
  documentation and/or other materials provided with the distribution.

- Neither the name of the Xiph.org Foundation nor the names of its
  contributors may be used to endorse or promote products derived from
  this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

TwistedWaveReference Manual