ReazonSpeech is a project to maintain freely-available Japanese audio datasets and ML models.
reazonspeech-espnet-next is a "bleeding-edge" repository that contains
latest ASR models trained by ReazonSpeech team.
We maintain this repository because we want to make our latest research results readily available, and try to incorporate feedback from community as quickly as possible.
If you are curious about the project, or have built something based on ReazonSpeech model, please let us know!
Project site: https://research.reazon.jp/projects/ReazonSpeech/index.html
- Downloads last month
This model can be loaded on the Inference API on-demand.