search

k2-fsa / sherpa-ncnn

Real-time speech recognition and voice activity detection (VAD) using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Linux, macOS, Windows, Raspberry Pi, VisionFive2, LicheePi4A etc.
https://k2-fsa.github.io/sherpa/ncnn/index.html
Apache License 2.0
1.07k stars 159 forks source link

Recognition of random words by speech recognition from microphone with endpoint detection #364

Open mukherjeesougata opened 3 weeks ago

mukherjeesougata commented 3 weeks ago

I am using this script for building live speech recognition API. But it is recognizing some random words in between. I have attached a figure which shows random words are getting recognized at indices 2,4 and 10. No words have been uttered at this indices. sherpa-cnn_API-problem

csukuangfj commented 3 weeks ago

Could you use https://k2-fsa.github.io/sherpa/ncnn/pretrained_models/zipformer-transucer-models.html#csukuangfj-sherpa-ncnn-streaming-zipformer-en-2023-02-13-english for testing if you are speaking English?

Also, please have a look at http://github.com/k2-fsa/sherpa-onnx which supports more models.