I need Soniox STT in javascript ?
# ask-ai
a
I need Soniox STT in javascript ?
t
Thanks for reaching out, Sidharth. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
You're asking for Soniox STT in JavaScript — thanks for asking. Soniox STT is currently only available as a LiveKit Agents plugin for Python. JavaScript/Node.js plugin support for Soniox is not available. Options you can use now: • Call Soniox’s REST API directly from your JavaScript/Node app (make HTTP requests to Soniox endpoints). • Stream audio from your JavaScript client to a Python agent (using LiveKit WebRTC) that runs the Soniox plugin and returns transcripts. • Use a LiveKit-supported STT provider with JavaScript support instead (examples: Deepgram or OpenAI). Which option would you like help with? • If REST API: I can provide a Node.js example showing how to send audio and receive transcripts. • If Python agent relay: I can outline the architecture and steps to stream audio from your client to the Python agent. • If switching providers: I can show a quick Deepgram/OpenAI Node.js example. Sources: Soniox STT plugin guide | LiveKit docs | Speech-to-text (STT) models | LiveKit Docs
a
I dont want a separate python and JS backend. I want to use Livekit JS SDK only.
t
Thanks — I hear you: you want to use only the LiveKit JavaScript SDK (no separate Python backend) and have Soniox STT in JS. Short answer: Soniox STT is not available as a LiveKit JS plugin. You cannot run the Soniox LiveKit plugin inside a pure-JS-only LiveKit agent. Options you can take while staying JS-only: • Call Soniox’s REST API directly from your Node.js app (send audio files or chunks over HTTP). I can provide a Node.js example for that. • Use a different STT provider that has official Node/JS support in the LiveKit ecosystem (examples: Deepgram or OpenAI). I can show how to wire one of those into a LiveKit JS flow. Which would you like: • A Node.js example calling Soniox REST for offline/HTTP transcription, or • A LiveKit JS example using Deepgram or OpenAI for real-time/streaming transcription?
a
So If I create a separate Soniox implementation can I pass the trasncription to livekit and it can pick up the flow from there?
t
Thanks — I hear you want to implement Soniox separately and have LiveKit "pick up" the flow. Short answer: you cannot inject external Soniox transcriptions into a LiveKit AgentSession to resume the agent's internal STT→LLM→TTS pipeline. What you can do (pick one and I’ll help next): • Send Soniox transcripts into the LiveKit room as text (e.g., sendText on topic
lk.transcription
) for UI display or client-side handling. This does not trigger agent logic. • Run a minimal LiveKit agent backend that uses the Soniox plugin—this is the supported way to keep the full agent workflow (STT → LLM → TTS) intact. • Keep everything JS-only and use a JS-supported STT provider (Deepgram or OpenAI) so you can both transcribe and integrate with LiveKit without a Python backend. Which option do you want to proceed with? If you choose the first, I can show a Node.js example to send transcripts into the room; if the second, I can outline the minimal agent setup; if the third, I can provide a LiveKit + Deepgram/OpenAI JS example. Sources: Soniox STT plugin guide | LiveKit docs | Transcriber | LiveKit docs | Text and transcriptions | LiveKit docs | Sending text | LiveKit docs