kjw11/Speaker-Aware-CTC

Speaker-aware CTC (SACTC) for multi-talker overlapped speech recognition.

28
/ 100
Experimental

This project offers an advanced method to accurately transcribe speech when multiple people are talking at the same time. It takes in audio recordings with overlapping speech and outputs distinct transcriptions for each speaker, even when they're speaking over each other. This is ideal for speech recognition researchers and engineers working on complex multi-speaker scenarios.

No commits in the last 6 months.

Use this if you are developing automatic speech recognition (ASR) systems and need to improve accuracy when multiple speakers are talking simultaneously.

Not ideal if you are looking for a standalone, ready-to-use application, as this is a core algorithm requiring integration with existing speech toolkits like ESPnet.

multi-speaker speech recognition speech separation audio transcription speaker diarization ASR research
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 4 / 25

How are scores calculated?

Stars

22

Forks

1

Language

Python

License

MIT

Last pushed

May 26, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/kjw11/Speaker-Aware-CTC"

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