3.11.0

Features

• The SDK now allows joining a conference as a mixed listener who receives one mixed audio stream and one mixed video stream from a conference. The feature lets up to 60,000 mixed listeners join a single conference. To join a conference as a mixed listener call the listen method and set the listener type to mixed.

• The SDK now allows sharing two screens in one conference, so two participants can share their screens at the same time.

• Improved the quality of voice fonts by increasing their sample rate to 48kHz.

• Improved noise reduction. The high noise reduction level now uses a Deep Neural Network (DNN) to remove all background sounds from your audio.

• Added new methods that allow the local participant to set the output audio volume. The setOutputVolume method sets the volume of all remote participants. The setVolume methods set the volume of a selected remote participant, audio track, or media stream in non-Dolby Voice conferences.

3.10.1

Changes

• Introduced a new stop method that lets you stop recording and playing audio samples. The method replaces the existing cancel method.

• Renamed the RecorderStatus model to AudioPreviewStatus. The existing recorderStatus model is still available, but it is deprecated.

Bug Fixes

• Fixed an issue where closing a set timer in the iOS Clock application resulted in not receiving remote participants' audio.
• Fixed an issue where recording audio samples during phone calls caused a crash.

Deprecated APIs

Deprecated the cancel method, which is now replaced with the stop method.

3.10.0

Features

Introduced a new preview method that allows the local participant to test different capture modes and voice fonts before a conference. The method sets a preview recorder that records the participant's audio and plays it back. Before playing the recorded audio, set the captureMode to a preferred setting that you wish to try.

Bug Fixes

Fixed an issue where some streamRemoved events were not correctly generated.

3.9.0

Features

• Introduced the ability to change CaptureMode in non-Dolby Voice conferences and use either the Standard or Unprocessed audio capture mode.

• Introduced a new voiceFont parameter to the Standard audio capture mode. Voice fonts allow participants to modify their voices in real time to improve social interaction in entertainment scenarios. For more information about this feature, see the Using Voice Fonts guide.

• Introduced an option to pre-process the local audio stream before sending the stream through an open session. You may use it to implement your own voice effects or remove or bleep out curse words from the sent audio. To start using this feature, use the processor accessor and the process method. For more information about pre-processing audio, see the Pre-processing the Local Audio Stream guide.

• Introduced a new updateParticipantInfo method that lets participants modify their names and avatars while they are in a conference.

• The SDK now supports receiving two shared screens in one conference.

Changes

Changed the default value of the audio capture mode in non-Dolby Voice conferences to Standard with high noise reduction. This setting optimizes captured audio for speech by aggressively removing non-speech content. If you want to transmit non-stationary background sounds to a conference and create a more natural audio experience, you can set noise reduction to low, which offers a similar experience as the default setting in previous releases. If you wish to transmit non-voice audio to a conference as well and use input device setting without any processing, use the Unprocessed mode.

3.8.3

Bug Fixes

Fixed a crash that occurred while leaving a conference.

3.9.0-beta.1

Features

• Introduced the ability to change CaptureMode in non-Dolby Voice conferences.

• Introduced a new voiceFont parameter to the Standard audio capture mode. Voice fonts allow participants to modify their voices in real time to improve social interaction in entertainment scenarios. For more information about this feature, see the Using Voice Fonts guide.

• Introduced an option to pre-process the local audio stream before sending the stream through an open session. You may use it to implement your own voice effects or build speech recognition and remove or bleep out curse words from the sent audio. To start using this feature, use the processor accessor and the process method.

Changes

• Changed the default value of the audio capture mode to Standard with high noise reduction. This setting optimizes captured audio for speech by aggressively removing non-speech content. If you want to transmit non-stationary background sounds to a conference and create a more realistic audio experience, you can set noise reduction to low. If you wish to transmit non-voice audio to a conference as well, use the Unprocessed mode.

3.8.2

Features

The SDK is now able to automatically reconnect the local participant to the current conference after momentary internet drops. When the local participant experiences a poor internet connection and gets disconnected from the conference, the SDK will wait for an internet connection to be stable again and reconnect the participant to the conference. If a participant shares a screen, a file, or a video before being disconnected, the shared media will not be started automatically after reconnection. If all remote participants leave the conference while the local participant is disconnected, the reconnection to the destroyed conference will fail.

3.8.2-beta.1

Features

The SDK is now able to automatically reconnect the local participant to the current conference after momentary internet drops. When the local participant experiences a poor internet connection and gets disconnected from the conference, the SDK will wait for an internet connection to be stable again and reconnect the participant to the conference. If a participant shares a screen, a file, or a video before being disconnected, the shared media will not be started automatically after reconnection. If all remote participants leave the conference while the local participant is disconnected, the reconnection to the destroyed conference will fail.

3.8.1

Bug Fixes

Fixed an issue where participants who used iOS devices could not share their screens with high resolution. The fix requires version 1.0.12 of the voxeet-screensharekit-ios package.

3.8.0

Features

• Introduced a new AudioSamples model that contains information about audio captured by the local participant's microphone, such as the sample rate, the audio format, and the buffer list. It allows you to access the participant's audio stream before sending the stream through an open session. You may use it to build speech recognition or audio moderation. To get the samples, use the delegate accessor and the audioRecordSamplesReady callback.

• Upgraded the WebRTC version to a recent version that contains improvements and bug fixes. The previously used version contains serious security vulnerabilities that can make your application susceptible to remote code execution, and can potentially give an attacker access to your application’s private data.

• Reduced the size of the SDK package.

Bug Fixes

• Fixed an issue where joining and leaving conferences caused freezing animations played on the main thread.
• Fixed an issue where a camera preview was not correctly rotated in landscape mode.