search

k2-fsa / sherpa-ncnn

Real-time speech recognition using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Raspberry Pi, VisionFive2, LicheePi4A etc.
https://k2-fsa.github.io/sherpa/ncnn/index.html
Apache License 2.0
891 stars 138 forks source link

ncnn带端点检测的流式语音识别bug #322

Open 20246688 opened 4 months ago

20246688 commented 4 months ago

在使用python-api-examples/speech-recognition-from-microphone-with-endpoint-detection.py的时候，一些词语的尾音会重复输出的怎么回事呢？是否可以处理？ 输出效果如下： Started! Please speak 0:重要 1:要 2:要 3:非常重要 4:要 5:要 6:没有问题 7:重要 8:要 9:要

csukuangfj commented 4 months ago

https://github.com/k2-fsa/sherpa-ncnn/blob/884ce6d9ebaf115ae80a2bb61ec8fe16bd999fc7/sherpa-ncnn/csrc/recognizer.cc#L216

你把这行删掉，然后再次编译，然后再测试下？

20246688 commented 4 months ago

删除后用git clone https://github.com/k2-fsa/sherpa-ncnn cd sherpa-ncnn python setup.py install吗？ 输出效果：Started! Please speak 0:重要 1:要 2:要 3:非常的重要 4:要 5:要 6:你 7:车

csukuangfj commented 4 months ago 
git clone https://github.com/k2-fsa/sherpa-ncnn
cd sherpa-ncnn

然后删除那一行

然后

python3 setup.py install
20246688 commented 4 months ago

是的结果还是一样的怎么办

csukuangfj commented 4 months ago

看看git diff的输出。你有重新编译吗

20246688 commented 4 months ago

啥意思？QQ上问吧