coval-ai / benchmarks
Reproducible voice-AI benchmarking — TTS / STT / S2S latency and accuracy.
Why this job: says: tts; topics: text-to-speech, speech-to-text, voice-ai
What it does
Most voice AI benchmarks are a one-time snapshot on clean studio audio. We run ours every 30 minutes, against almost 40 providers, on audio that sounds like real calls: accents, background noise, echoey rooms, far-away mics, clipped signals, and phone-line compression.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit today
- Stars
- 105
- On GitHub
- Licence
- Custom
- Read before business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Python
- 100% of the code
Is it looked after?
- Contributors
- 6
- Last 90 days
- Issues
- 69% closed
- 5 still open
- Pull requests
- 757 merged
- 0 waiting
- Releases
- None
- Never published
- Contributing guide (opens in new tab)
- Security policy (opens in new tab)
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| SaynaAI / sayna | 317 | 5 months ago | Apache-2.0 | Mac, Windows, Linux |
| QwenAudio / qwen-audio-agent | 3,191 | 5 days ago | Apache-2.0 | Mac, Windows |
| deepgram / deepgram-python-sdk | 480 | today | MIT | No downloads |
More: languages, topics and project record
Languages
- Python 100%
- Shell <1%
- Dockerfile <1%
- Mako <1%
Topics
- ai
- ai-agents
- speech-to-text
- text-to-speech
- voice-ai
- agent-evaluation
- ai-evals
- evals
- model-comparison
- speech-to-speech
- voice-agent
- voice-agents
Project record
- JobSpeech and audio
- StatusListed
- Commits804
- Forks15
- Watching4
- Stars in 28 days+3
- CreatedApr 29, 2026
- First seen by usNov 7, 2026
- CheckedNov 10, 2026