Looking for real-time STT that can auto-detect language on the fly (no language hint at init) for a voice app. Ideal if it handles mid-utterance code-switching (e.g., Spanish↔️English) with low latency. What providers/models are people using with LiveKit today? i am currently using grop whisper large
The best bet in these use cases is the realtime api's you can try OpenAI, gemini in real-time options
If you are looking for an STT provider specifically, you can look at Nova-3 ( deepgram ) or universal-model by assembly ai