Menu
Streaming text-to-speech via OpenAI Audio Speech API with adaptive sentence chunking
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Edit, transform, and enhance images using AI models
Extract text from images using OCR and vision AI
Neural text-to-speech via Amazon Polly
Speaker diarization โ identifies and tracks who is speaking at each moment in an audio stream
Semantic endpoint detection โ uses an LLM to classify whether the user's utterance is a complete thought, reducing false turn boundaries on mid-sentence pauses
Batch speech-to-text via Google Cloud Speech-to-Text API
| name | streaming-tts-openai |
| description | Streaming text-to-speech via OpenAI Audio Speech API with adaptive sentence chunking |
| category | voice |
Low-latency streaming text-to-speech using OpenAI's TTS API. Buffers incoming LLM tokens into natural sentence chunks before making API requests, enabling audio playback to begin within the first sentence rather than waiting for full LLM output.
Set OPENAI_API_KEY in your environment or agent secrets store.
., ?, !, or ; boundaryIn agent.config.json:
{
"voice": {
"tts": "openai"
}
}
Provider-specific options via providerOptions:
{
"voice": {
"tts": "openai",
"providerOptions": {
"model": "tts-1",
"voice": "nova",
"format": "opus",
"maxBufferMs": 2000
}
}
}
| Event | Payload | Description |
|---|---|---|
utterance_start | { text: string } | Sentence chunk dispatched for synthesis |
audio_chunk | EncodedAudioChunk | Synthesized audio buffer ready for playback |
utterance_complete | { text, durationMs } | Synthesis complete for a sentence chunk |
cancelled | { remaining: string } | Session was cancelled; remaining text not rendered |
error | Error | Synthesis request failed |
close | โ | Session fully terminated |