KoljaB/RealtimeSTT
A Python library for real-time speech-to-text transcription using ML models like Whisper.
Not currently ranked — collecting fresh signals.
star history
RealtimeSTT provides low-latency speech-to-text with voice activity detection, wake word activation, and instant transcription. It primarily uses faster-whisper as its backend and supports multiple ASR engines including kroko_onnx for local streaming. The library integrates Silero VAD for voice activity detection and offers direct audio stream access for developers building voice-enabled applications.
Frequently asked
- What is KoljaB/RealtimeSTT?
- A Python library for real-time speech-to-text transcription using ML models like Whisper.
- Is RealtimeSTT open source?
- Yes — KoljaB/RealtimeSTT is open source, released under the MIT license.
- What language is RealtimeSTT written in?
- KoljaB/RealtimeSTT is primarily written in Python.
- How popular is RealtimeSTT?
- KoljaB/RealtimeSTT has 9.9k stars on GitHub.
- Where can I find RealtimeSTT?
- KoljaB/RealtimeSTT is on GitHub at https://github.com/KoljaB/RealtimeSTT.