Stream TTS audio with bufferStream

Now that we can synthesize a full utterance, we're going to surface the audio as it synthesizes. For low-latency voice, the reader wants each chunk as soon as the engine produces it.

textToSpeech({ stream: true }) returns { buffer, bufferStream, done } instead of just { buffer }. The samples live on result.bufferStream as an AsyncGenerator<number>.

Setting stream: true flips the result to { buffer, bufferStream, done }. You would call it like so:

const result = textToSpeech({
  modelId,
  text: "Streaming chunks as the engine synthesizes them is the right latency for low-latency voice.",
  inputType: "text",
  stream: true,
});

The canonical way to consume samples is iterating result.bufferStream with for await:

let totalSamples = 0;
for await (const sample of result.bufferStream) {
  void sample;
  totalSamples += 1;
}
console.log(`▸ Streamed ${totalSamples} samples`);

Each number is a single PCM sample. Iterating with for await produces samples in the order they were synthesized. result.buffer is empty when stream: true; the samples live on the generator instead.

result.done is a Promise<boolean> that resolves when synthesis finishes. Await it from a separate branch if you need to know when the stream terminates.

Note: this is different from textToSpeechStream, which is a duplex session for piping tokens from a streaming LLM into TTS. Use textToSpeech({ stream: true }) when you already have the full text and want a stream of audio chunks.