effect-uai streaming-synthesis

Incremental text in, audio chunks out. Use SpeechSynthesizer.streamSynthesisFrom when the text is still being written and the user should already hear the first phrase.

Reach for this when the user says any of:

• "Read LLM output aloud as it streams"
• "Pipe token deltas into TTS"
• "Start TTS playback before the full answer exists"

Pattern

import { Stream } from "effect"
import * as SpeechSynthesizer from "@effect-uai/core/SpeechSynthesizer"