reazon-research/ReazonSpeech

Massive open Japanese speech corpus

52
/ 100
Established

This project provides advanced Japanese speech recognition capabilities, converting spoken Japanese audio into text. It takes audio recordings, including noisy ones, and outputs accurate text transcripts. This is ideal for researchers or developers creating applications that need to understand spoken Japanese, especially in human-robot interaction or other noisy environments.

373 stars.

Use this if you need to accurately transcribe Japanese audio, even if it's mixed with English or contains background noise, for applications like voice assistants or robotics.

Not ideal if your primary need is for speech recognition in languages other than Japanese, or if you require an off-the-shelf, no-code transcription service.

Japanese speech-to-text Human-robot interaction Audio transcription Multimodal speech recognition Speech corpus development
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

373

Forks

34

Language

Python

License

Apache-2.0

Last pushed

Jan 19, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/reazon-research/ReazonSpeech"

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