Table of Contents

Media

The media layer moves audio between calls, devices and your code. client.Media (MediaManager) is the factory for every media primitive.

Primitives

IMediaReceiver receiver = client.Media.CreateReceiver();  // pull encoded frames out
IMediaSender   sender   = client.Media.CreateSender();    // push frames into a call
MediaConnector connector = client.Media.CreateConnector(); // bridge two calls
  • IMediaReceiver — receives encoded MediaFrames (payload in the negotiated codec) from a call (for transcription, streaming to an AI backend, custom mixing).
  • IMediaSender — SendAsync(MediaFrame) injects audio into a call (TTS, prompts, synthesized speech).
  • MediaConnector — cross-connects two active calls into a bridge.

Attaching these to a call is the media tap; bridging is the bridge guide.

Video

Video has its own transport-only primitives — the SDK moves encoded frames but never encodes or decodes:

IVideoReceiver videoIn  = client.Media.CreateVideoReceiver(); // inbound encoded VideoFrames
IVideoSender   videoOut = client.Media.CreateVideoSender();   // outbound encoded VideoFrames

IVideoSender also surfaces the SDK's recommended outbound bitrate and peer keyframe requests so you can drive your encoder. Bring your own codec (VP8, H.264, …). Full walkthrough: Video calls.

Default audio device

For a plain softphone you usually skip the primitives and let the SDK wire the OS device:

await client.AttachDefaultAudioAsync(call);
await client.DetachDefaultAudioAsync(call);

Runtime device switching, mute and volume live on the client — see Audio devices.

Recording and playback

IRecordingSession rec = await client.Media.StartCallRecordingAsync(call, "call.wav");
await rec.PauseAsync(); await rec.ResumeAsync(); await rec.StopAsync();

IPlaybackSession play = await client.Media.StartCallPlaybackAsync(call, "prompt.wav");
await play.StopAsync();

Conference variants (StartConferenceRecordingAsync / StartConferencePlaybackAsync) operate on a bridge. Details: Recording/playback.

Codecs and encryption

Codec preference is set once via VoipConfiguration.PreferredAudioCodecs; SRTP/SRTCP is negotiated automatically per SrtpPolicy. Neither changes the media-primitive API — you work with MediaFrames whose payload is encoded in the negotiated codec (decode/encode it yourself); SRTP/SRTCP and transport stay transparent.

Edit this page