sherpa-onnx

sherpa-onnx copied to clipboard

[WIP] Encode hotwords in C++ side

2

pkufool

使用 sherpa-onnx-streaming-zipformer-multi-zh-hans-2023-12-12 模型，热词异常

5

使用命令行生成中文字热词文件后，在tokens.txt能找到对应的byte，但是使用时有异常   

jianking123

sherpa-onnx-keyword-spotter 使用命令行工具 识别结果找不到

1

按照[kws文档说明](https://k2-fsa.github.io/sherpa/onnx/kws/index.html)的指引 `sherpa-onnx-keyword-spotter --help`，使用命令行工具执行语音唤醒脚本 貌似输出只有配置信息，没有看到识别结果，但是执行的脚本是编译后的格式，无法检查或者print日志信息。 不知道下一步怎么做了。 执行： ``` ./build/bin/sherpa-onnx-keyword-spotter \ --tokens=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/tokens.txt \ --encoder=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/encoder-epoch-12-avg-2-chunk-16-left-64.onnx \ --decoder=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/decoder-epoch-12-avg-2-chunk-16-left-64.onnx \ --joiner=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/joiner-epoch-12-avg-2-chunk-16-left-64.onnx \ --provider=cpu \ --num-threads=2 \ --keywords-file=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/test_wavs/test_keywords.txt \ sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/test_wavs/0.wav ``` 输出日志： ``` /home/taorui/sherpa-onnx/sherpa-onnx/csrc/parse-options.cc:Read:361 ./build/bin/sherpa-onnx-keyword-spotter --tokens=sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/tokens.txt...

taorui-plus

Decoding method 'modified_beam_search' gives letters/words on silence, while 'greedy_search' works well

17

When I set decoding method to 'modified_beam_search' it returns letters/words on silence, after short time, periodically. But when I use default 'greedy_search' decoding method everything works well. As an example...

ChrystianKacki

TTS WebAssembly for other languages not work

10

I tried to follow the instruction to build text-to-speech with WebAssembly. https://k2-fsa.github.io/sherpa/onnx/tts/wasm/index.html When I used English language as in instruction. It worked well. But when I tried to use some...

kmpartner

sherpa-onnx-offline-tts: Allow piping text using cat or text file input with command line parameter

6

Hi all, Is it possible to add the possibility to pipe text into the tts engine using cat or a command line parameter? That would help greatly when converting large...

domasofan

TTS: Add a SAPI 5 driver for Windows

3

Hi, That would be interesting and very beneficial for Windows users of screenreaders and other speech based software. Adding a SAPI 5 driver for text to speech to interface with...

domasofan

Android TTS: Some ideas for packaging tts engine and voices/models

2

Hi all, Just tested the tts engine on android 14 on a phone with an 64 bit processor. I also tested tts with the talkback screenreader. It works very well...

domasofan

tts: Increase TTS volume to 0 DB.

2

Hi all, This seems to affect tts in general. Tested on Android 14 and Windows 11. the generated speech seems to be very quiet and it seems to be under...

domasofan

Hi all, It would be nice if it would be possible to add this feature so audio can be piped directly for example into ffmpeg or lame. So generating a...

domasofan