ictnlp/StreamSpeech

StreamSpeech is an “All in One” seamless model for offline and simultaneous speech recognition, speech translation and speech synthesis.

39
/ 100
Emerging

Built on a unified encoder-decoder architecture trained with multi-task learning, it handles ASR, speech-to-text translation, speech-to-speech translation, and TTS in both offline and low-latency streaming modes using fairseq. The model generates intermediate results incrementally during simultaneous translation, enabling real-time ASR transcriptions and translation outputs before final speech synthesis completes.

1,252 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 9 / 25
Community 18 / 25

How are scores calculated?

Stars

1,252

Forks

102

Language

Python

License

MIT

Last pushed

Jun 29, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/ictnlp/StreamSpeech"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.