Table of Contents

Warning : Some protected information on this page is not displayed.
Ensure you are logged in if you are a licensed user for specific platforms.

Creating Audio Conversion Settings ShareSets

The ShareSets you create to manage the Conversion Settings are based on the needs of your project and the requirements of each active platform. Many of your choices here can have a big impact on the performance and quality of your audio project. After applying the ShareSets to the objects in your project, you can go back and tweak your conversion setting ShareSets at any time to achieve the best possible quality within the constraints of the platform and the game. When you import audio files you can also speed up the process by re-using the same ShareSets that you have defined here for language and source versions.

[Note] Note

You can use one of your ShareSets as the default Conversion Settings for your project. For more information on setting up the default Conversion Settings for your project, refer to Specifying the Default Conversion Settings.

The Conversion Settings Editor is divided into two main sections:

  • Settings: The area above the Audio Sources. It allows you to set the Conversion Settings for each platform, including sample rate, audio format, and number of channels.

  • Results: The area listing all your audio sources. It allows you to compare the original and converted sources, including the number of channels, sample rate, and file size.

The audio conversion process retains the same pitch and duration as the original files; however, you can define the following properties for your conversion:

You can also specify whether you want to:

  • Insert a filename marker for lip-syncing or sub-titling;

  • Remove DC offset;

    [Caution] Caution

    We recommend not removing the DC offset for looping sounds. The remove mechanism is a high-pass filter, so there is no guarantee that it will modify the first and last sample of the loop in the same way because it does not know they will play contiguously. This could create a discontinuity in the signal, which is audible as a click.

  • Apply dither; or

  • Allow channel upmix, which means mono source files will be converted to stereo when channels are marked as Stereo or Stereo drop.

    [Note] Note

    If this option is not selected, mono source files will remain mono regardless of the Channels settings.

About Audio Channels

When converting multi-channel audio sources, you must decide which channels to preserve. For more information, refer to Channel configuration.

In Wwise, you can choose from the following options:

Channel Option

Description

Comments

As Input

Preserves the same number of audio channels as the original media file.

Some audio formats may not be supported on some platforms, in which case multichannel files will be downmixed to stereo.

Mono

All channels are mixed into one.

The L-R Mix is only used for Stereo to Mono conversion. Other channel configurations are downmixed according to Appendix A, Downmixing Behavior.

The LFE channel is always dropped.

Mono drop

All channels are dropped except the first one.

Depending on the channel configuration of the original file, the first channel can either be the Left or Center channel.

Stereo

All channels are mixed into front left and front right.

The L-R Mix is only used for Mono to Stereo conversion. Other channel configurations are downmixed according to Appendix A, Downmixing Behavior.

The LFE channel is always dropped.

Stereo drop

All channels are dropped except channels that are defined as Left or Right.

In cases where no channel is defined as Left or Right, and there is a Center channel defined (mono), the following conversion occurs:

Left = 0.707C Right = 0.707C

The converted file will be twice the size of the original file.

It is important to note that Wwise does not do any multi-channel encoding; it simply feeds LPCM data to the console or system in either stereo, or 5.1 or 7.1 surround. Once the LPCM data is received by the console or system, it can then be output in almost any format supported by the particular console or system, including Dolby, DTS, or DPL2. Some restrictions do exist, however, including:

  • Support for only stereo output on the Android and iOS platforms.

  • Support for only stereo and 5.1 surround output on the Switch™ platform.

  • Support for only stereo, and 5.1 and 7.1 surround output on the Mac and tvOS platforms.

  • Other platforms, such as Windows, PlayStation 4, and Xbox One, natively support up to 7.1 channels at their output. Wwise is able to carry all standard channel configurations (up to 13.1 channels), as well as anonymous configurations (up to 256 channels). Note that these configurations require the use of a special sink plug-in that can interpret them and/or pass them to dedicated hardware.

About sample rates

The sample rate determines the number of times per second a digital audio signal is sampled. When deciding on which sample rate to choose, many factors come into play and like other quality/performance issues, setting the sample rate is a balancing act. To give you as much control as possible, Wwise gives you many different sample rate conversion options:

  • As Input - Converts the file using the same sample rate as the original file. If the sample rate is not available for a particular platform or audio format, then the closest sample rate available will be used.

  • Auto (Low/Medium/High) - Converts the file using a sample rate selected by Wwise after performing an FFT analysis of the file. The difference between the low, medium, and high quality settings is the cutoff threshold value used by the algorithm. You can tweak the quality level of each setting by defining their threshold values in the Project Settings dialog box. For more information on the automatic sample rate detection performed by Wwise, refer to Defining the Sample Rate Automatic Detection Settings.

  • 300 to 48,000 - Converts the file using a specific sample rate. The sample rate range varies for each platform up to a maximum of 48,000 Hz.

[Note] Note

In the cases As Input and Auto, you can further restrict the sample rate via the Min Sample Rate and Max Sample Rate combo boxes.

About Audio Formats

Before converting your audio files, you need to decide into which format they will be converted. Wwise supports multiple different audio formats to give you greater flexibility and control to work around the limitations of each platform.

Due to the different specifications of each platform, not all formats are available on each one. The following table displays which audio formats are available by platform.

Platform

AAC

ADPCM

Opus (standard)

PCM

Vorbis

WEM Opus

Android

 

iOS

Linux

 

Mac

Switch

 

PS4

 

Windows

 

Xbox One

 

Platform-specific audio formats supported by Wwise:


[Note] Note

Only ADPCM, Opus (standard), WEM Opus, PCM, and Vorbis audio formats support all configurations of multi-channel files. When the channel configuration is not supported by the format, Wwise downmixes it to the next supported configuration.


Here are short descriptions of each of the different audio formats supported by Wwise:

  • AAC - A perceptual coding audio compression method for the Mac and iOS. AAC is said to achieve better sound quality than MP3 at similar bit rates. Compression is variable, content dependent, and the quality setting can be controlled by the “quality” slider. On iOS, AAC is decoded by the hardware-assisted codec if it is available. Note that the hardware may only decode one AAC sound at a time.

    [Note] Note

    Because of the presence of hardware support for AAC on iOS, this format is recommended for your background music on this platform. However, you should ensure that there is never more than one AAC sound playing at the same time. Additional AAC sounds are decoded in software, which uses a huge amount of CPU. Thus, you should avoid using this format with interactive music, since Music Segments usually overlap. Note that when using hardware-assisted decoding, the CPU usage displayed in the Wwise Profiler will appear to be very large. This is because decoding occurs within the audio processing thread of Wwise. On the other hand, most of this time is actually spent waiting for the hardware, during which the CPU is available to other threads of your game. Also, starting a new AAC voice may take a significant amount of time, and may even cause voice starvation on iOS. AAC has no limitation on looping, but it is not appropriate for sample-accurate playback. Only the following channel configurations are allowed (other configurations are automatically downmixed to stereo): mono, stereo, 0.1, 1.1, 5.0 and 5.1. Note that multichannel files and the LFE channel are not supported on iOS.

  • ADPCM - An audio file conversion encoding method that quantizes the difference between a sound signal and a prediction that has been made of that sound signal. The ADPCM quantization step is adaptive, which differs from PCM encoding where the signals are quantized directly. Fundamentally, ADPCM offers significant gains in storage and CPU usage at the cost of sound quality. As such, it is typically used on mobile platforms.

  • Opus (standard) - A low-latency audio codec, optimized for both voice and general-purpose audio, which outperforms other codecs for compression without compromising on sound quality. The balance between data compression efficiency and perceived sound quality is controlled using the Bitrate setting; the higher the rate, the better the sound quality is. This version is a standard implementation of the Opus specification, supported on all platforms. The Opus decoder needs a lot of pre-roll to start decoding, which incurs a significant cost in CPU and streaming resources. Therefore, it is not suited for applications that need a lot of seeking or looping, such as in complex Interactive Music, sample-accurate loops, or trigger-rate containers.

    [Note] Note

    This format is deprecated and may be removed from Wwise in the near future. It is strongly recommended that you use WEM Opus instead.

  • WEM Opus - Audio encoding is identical to standard Opus, but stored in a container that is better optimized for seeking and looping when streamed from disk. Unlike standard Opus, decoding is hardware-accelerated on some platforms, thus circumventing the format's greatest disadvantage; CPU usage.

    This format intends to completely replace the standard Opus format. You should always choose this format over standard Opus.

  • PCM - An audio file conversion encoding method where distinct binary representations or pulse codes are chosen. These are then quantized by measuring values between two encoded points, selecting the value associated with the nearest point.

  • Vorbis - A perceptual encoding method that permits encoding of audio files at fixed and variable bitrates while maintaining a very good perceived sound quality. The balance between data compression efficiency and perceived sound quality is controlled using the Quality Factor setting or by specifying the maximum, minimum, and average bitrates per channel. The Vorbis encoder may require the use of a seek table. For more information, refer to Using seek tables with the Vorbis encoder.

    [Note] Note

    Audiokinetic's special implementation of Vorbis has been highly optimized for all platforms.

Each format has its advantages and disadvantages and the format you choose will depend on the CPU and memory restrictions of your particular game. For a further discussion on when to use which audio format, refer to Audio formats.

About DC Offsets

It is a good idea to remove the DC offset using a DC offset filter because DC offsets can affect volume and cause artifacts in Wwise. There are some cases, however, where you should not remove the DC offset, for example, for sample accurate containers. In other cases, for example, where sounds are normalized to 0 dB, you may or may not need to remove the DC offset. During the conversion process, DC offsets are removed by default. You can, however, disable this setting if needed in the Conversion Settings dialog box.

[Caution] Caution

If you are generating motion directly from your audio sources, you should be aware that the removal of the DC offset may alter the motion output for controllers.

To create a Conversion Settings ShareSet:

  1. In the Project Explorer, switch to the ShareSets tab.

  2. In the Conversion Settings section, select the Work Unit into which you want to create the new ShareSet.

  3. From the Project Explorer toolbar, click the Conversion Settings icon .

    A new Conversion Settings ShareSet is created in the selected Work Unit.

  4. Name the ShareSet appropriately and press Enter.

  5. Double-click the new ShareSet to load it in the Conversion Settings Editor.

  6. Specify the Channels for each platform by selecting one of the following:

    • As Input - To preserve the same number of audio channels as the original media file.

    • Mono - To mix all channels into one mono channel.

    • Mono drop - To drop all channels except the first one.

    • Stereo - To mix all channels into the front left and front right.

    • Stereo drop - To drop all channels except the ones defined as Left and Right.

    [Note] Note

    By default, the channel settings are linked across all platforms. To specify a unique channel setting for a particular platform, unlink the property first and then define the setting.

    [Note] Note

    If you don't want to increase the number of channels for mono files, make sure to disable the Allow channel upmix option.

  7. If you are converting a stereo source to mono or vice versa, you can use the L-R Mix settings to specify the power level of the signal assigned to the left and right channels.

    [Note] Note

    By default, the L-R Mix settings are unlinked across all platforms. To specify a common L-R Mix setting for particular platforms, link the property first and then define the setting.

  8. From the Sample Rate list, select the frequency with which the audio files will be sampled per second during conversion. Depending on the particular circumstances of your game, you can select one of the following options:

    • As Input - To convert the file using the same sample rate as the original file. If the sample rate is not available for a particular platform or audio format, then the closest sample rate available will be used.

    • Auto (Low/Medium/High) - To convert the file using a sample rate selected by Wwise after performing an FFT analysis of the file. The difference between the low, medium, and high quality settings is the cutoff threshold value that identifies the frequency used to determine the best sample rate in which to convert your files. You can tweak the quality level of each setting by defining their threshold values in the Project Settings dialog box. For more information on the automatic sample rate detection performed by Wwise, refer to Defining the Sample Rate Automatic Detection Settings.

    • 300 to 48,000 - To convert the file using a specific sample rate. The sample rate range varies for each platform up to a maximum of 48,000 Hz.

  9. If the Sample Rate is set to either As Input or Auto, then use the Min Sample Rate and Max Sample Rate entries to restrict the conversion sampling rate.

  10. Specify the audio format for the conversion by selecting one of the following: AAC, ADPCM, ATRAC9, OPUS, OPUSNX, PCM, Vorbis, or XMA.

    [Note] Note

    Click the Edit button to modify the encoding parameters for the AAC, ADPCM, Opus, Vorbis, and XMA audio formats. For a complete description about the encoding parameters for these audio formats, click the Help button in the corresponding dialog box. For information on best practices for selecting parameters for the Vorbis audio format, refer to the Vorbis Encoder Parameters page in the reference documentation.

  11. From the Sample rate conversion quality list, select the method that will be used to convert the file's sample rate. You can select either of the following options:

    • Normal (Faster) - Produces a good quality conversion that is three to six times faster than the Best option.

    • High (Slower) - Produces the best quality conversion.

    [Note] Note

    If you expect your content to contain high frequencies and you are converting to sample rates below 24 kHz, it is recommended that you use the High option.

  12. If you want a marker to be created at the beginning of each converted file, select Yes from the Insert Filename Marker list.

    The marker will only contain the filename and not the file's path and extension. Having the name visible can be useful when you want to bind an action to a sound playing in the sound engine, for example, when lip-syncing or sub-titling.

  13. If you don't want DC offset to be removed during the conversion process, clear the Remove DC Offset check box.

    By default this option is selected. In most cases, it is preferable to remove any DC offset. There are cases, however, where the DC offset may not need to be removed, including:

    • Sounds that will be added to sample accurate containers.

    • Sounds that are normalized to 0 dB.

      For more information about how DC offsets affect audio signals in Wwise, refer to Removing DC Offsets.

    [Caution] Caution

    If you are generating motion directly from your audio sources, you should be aware that the removal of the DC offset will alter the motion output.

  14. If you don't want dithering to be applied during the bit rate conversion, clear the Apply Dither check box.

    Dither is the noise added to a signal prior to quantization in order to reduce the distortion and noise modulation that results from the quantization process. Dithering is only applied when the resolution changes, for example, from 24 bits to 16.

  15. Close the Conversion Settings Editor.

    The settings that you have specified are automatically saved and the ShareSet can now be assigned to one or more objects in your project hierarchy.

  16. Repeat steps 1-14 to create as many Conversion Settings ShareSets as needed for your project.

    [Note] Note

    Before the Audio Sources table is populated, you must assign the Conversion Settings ShareSet to an object and then convert the audio files using these settings. For more information on assigning a Conversion Settings ShareSet to an object, refer to Assigning Conversion Settings ShareSets to Objects.