Voice-Streaming & Text-to-Speech (TTS) Latency

How does the App Intents framework support streaming spoken voice (TTS) output for long-form text responses? Is there an API (such as an asynchronous sequence or buffer stream) that allows Siri to begin speaking a response while the underlying AI engine is still generating the remainder of the text?

Voice-Streaming & Text-to-Speech (TTS) Latency
 
 
Q