How can I use japanese as the language for keyword spotting - Githubissues

k2-fsa / sherpa-onnx

Speech-to-text, text-to-speech, speaker recognition, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, Raspberry Pi, RISC-V, x86_64 servers, websocket server/client, C/C++, Python, Kotlin, C#, Go, NodeJS, Java, Swift, Dart, JavaScript, Flutter, Object Pascal, Lazarus, Rust

https://k2-fsa.github.io/sherpa/onnx/index.html

Apache License 2.0

3.11k stars 359 forks source link

How can I use japanese as the language for keyword spotting #1250

Open itsrealkk opened 4 weeks ago

itsrealkk commented 4 weeks ago

Hi, i have tried the Chinese wenet model, it's awesome.

Btw, I need a japanese model for kws. I have noticed that sherpa-onnx-zipformer-ja-reazonspeech-2024-08-01 model. However, this model is so large.

How can I train a 3.3M ja model and use japanese for kws

csukuangfj commented 4 weeks ago

Currently, we don't have a KWS model for Japanese.

You can train it by yourself.

csukuangfj commented 4 weeks ago

How can I train a 3.3M ja model and use japanese for kws

Please refer to how we train a KWS English model or KWS Chinese model in icefall.