Chat Completions Audio Responses

Gabber supports low-latency audio responses for chat completions.

This means you don't need to separately stream audio generation responses for chat/completion calls.

Requesting Voice

To request voice, an additional gabber field must be used in the completion request containing the Gabber Voice Id of interest.

const body = {
  model: selectedLlm.id,
  messages: [
    { role: "system", content: "Once upon a time" },
  ],
  stream: true,
  gabber: {
    voice: selectedVoice?.id,
  }
};

const responseStream = await openAI.chat.completions.create(
  { stream: true, model: "", messages: [] }, // To satisfy the OpenAI SDK typings
  { body },
);