all_inputs - Output will be terminated if all of the input streams are finished. In particular, output stream will be terminated if no inputs were ever connected.
type OutputAudioOptions = {
initial: Audio;
channels: "mono" | "stereo";
forward_error_correction?: bool;
encoder_preset?: "quality" | "voip" | "lowest_latency";
send_eos_when?: OutputEndCondition;
}
type OutputAudioOptions = {
initial: Audio;
channels: "mono" | "stereo";
forward_error_correction?: bool;
encoder_preset?: "quality" | "voip" | "lowest_latency";
mixing_strategy?: "SumClip" | "SumScale";
send_eos_when?: OutputEndCondition;
}
initial - Initial audio for output."lowest_latency" - Only use when lowest-achievable latency is what matters most.
mixing_strategy - (**default="sum_clip") Specifies how audio should be mixed.
+"SumClip" - Firstly, input samples are summed. If the result is outside the i16 PCM range, it gets clipped."SumScale" - Firstly, input samples are summed. If the result is outside the i16 PCM range, nearby summed samples are scaled down by factor, such that the summed wave is in the i16 PCM range.send_eos_when - Condition for termination of output stream based on the input streams states.send_eos_when - Defines when output stream should end if some of the input streams are finished. If output includes both audio and video streams, then EOS needs to be sent on both.initial - Root of a component tree/scene that should be rendered for the output. Use update_output request to update this value after registration. Learn more.type Audio = {
channels: "stereo" | "mono";
forward_error_correction?: boolean;
encoder_preset?: AudioEncoderPreset;
send_eos_when?: EosCondition;
initial: {
inputs: AudioInput[];
};
}
type AudioInput = {
input_id: string;
volume?: number;
}
type AudioEncoderPreset =
| "quality"
| "voip"
| "lowest_latency"
type Audio = {
channels: "stereo" | "mono";
forward_error_correction?: boolean;
encoder_preset?: AudioEncoderPreset;
send_eos_when?: EosCondition;
initial: {
inputs: AudioInput[];
};
mixing_strategy?: "sum_clip" | "sum_scale"
}
type AudioInput = {
input_id: string;
volume?: number;
}
type AudioEncoderPreset =
| "quality"
| "voip"
| "lowest_latency"
channels - Channel configuration for output audio.forward_error_correction - (default=false) Specifies whether the stream use forward error correction. It's specific for Opus codec. For more information, check out RFC.initial - Initial configuration for audio mixer for this output. Use update_output request to update this value after registration.initial.inputs[].input_id - Input ID.initial.inputs[].volume - (default=1.0) Float in [0, 1] range representing volume.mixing_strategy - (default=sum_clip) Specifies how input samples should be mixed:
+sum_clip - Firstly, input samples are summed. If the result sample is outside the i16 PCM range, it gets clipped.sum_scale - Firstly, input samples are summed. If the result wave is outside the i16 PCM range, nearby samples are scaled down by factor, such that the summed wave is in the i16 PCM range.type EosCondition = {
any_input?: bool;
all_inputs?: bool;
any_of?: InputId[];
all_of?: InputId[];
}
This type defines when end of an input stream should trigger end of the output stream. Only one of those fields can be set at the time.
diff --git a/docs/api/renderers/image.html b/docs/api/renderers/image.html index 133bdd8c3..df4c5f0a9 100644 --- a/docs/api/renderers/image.html +++ b/docs/api/renderers/image.html @@ -4,7 +4,7 @@