revdotcom/reverb
Open source inference and evaluation code for Rev's speech-to-text and speaker diarization models.
Not currently ranked — collecting fresh signals.
star history
Reverb provides open source inference code for Rev’s ASR and diarization models. The ASR component uses the WeNet framework and the speaker diarization component uses the Pyannote framework. Models are available on HuggingFace and benchmarks compare performance against Whisper and Canary on standard datasets.
Frequently asked
- What is revdotcom/reverb?
- Open source inference and evaluation code for Rev's speech-to-text and speaker diarization models.
- Is reverb open source?
- Yes — revdotcom/reverb is open source, released under the Apache-2.0 license.
- What language is reverb written in?
- revdotcom/reverb is primarily written in Python.
- How popular is reverb?
- revdotcom/reverb has 437 stars on GitHub.
- Where can I find reverb?
- revdotcom/reverb is on GitHub at https://github.com/revdotcom/reverb.