dimastatz/whisper-flow
Whisper-Flow is a Python framework for real-time speech-to-text transcription of streaming audio using OpenAI's Whisper model.
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history
Whisper-Flow accepts a continuous stream of audio chunks and produces incremental transcripts immediately, rather than processing entire files in batch mode. It applies temporal windowing techniques, such as tumbling windows, to transcribe streaming media on the fly. This enables applications to incorporate live speech-to-text functionality with faster transcription times, potentially at some cost in accuracy.
Frequently asked
- What is dimastatz/whisper-flow?
- Whisper-Flow is a Python framework for real-time speech-to-text transcription of streaming audio using OpenAI's Whisper model.
- Is whisper-flow open source?
- Yes — dimastatz/whisper-flow is open source, released under the MIT license.
- What language is whisper-flow written in?
- dimastatz/whisper-flow is primarily written in Python.
- How popular is whisper-flow?
- dimastatz/whisper-flow has 1k stars on GitHub.
- Where can I find whisper-flow?
- dimastatz/whisper-flow is on GitHub at https://github.com/dimastatz/whisper-flow.