FunASR
FunASR copied to clipboard
A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.
## 🐛 Bug 测试自己fintune训练得到的模型,在onnx推理时,当推理的测试音频中包含“@”时,有报错 经过调试定位到这个地方无法处理“@”字符,请问自己fintune训练时,对于@可以标注为@吗?(我目前是这样做的)我看官方推理的结果是“艾特” - OS (e.g., Linux): - FunASR Version (e.g., 1.0.25):
## 🐛 Bug When I execute model.generate() with an audio almost an hour, the error “list index out of range” appears. ### To Reproduce ``` rtf_avg: 0.511: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:23
#### What is your question? 使用 FunASR实时语音听写服务,进行实时音视频转文字听写,使用的是2pass模式 服务启动命令 ``` nohup bash run_server_2pass.sh \ --download-model-dir /workspace/models \ --vad-dir damo/speech_fsmn_vad_zh-cn-16k-common-onnx \ --model-dir damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-onnx \ --online-model-dir damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online-onnx \ --punc-dir damo/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727-onnx \ --lm-dir damo/speech_ngram_lm_zh-cn-ai-wesp-fst...
英文模型用的是paraformer-eng,使用AutoModel进行batch推理和单条推理。经测听,batch推理的结果错的更多,都是少字。 中文模型(speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch)的batch推理则正常,虽然与单条推理的结果也有差异,但两者的错误率差不多。
Notice: In order to resolve issues more efficiently, please raise issue following the template. (注意:为了更加高效率解决您遇到的问题,请按照模板提问,补充细节) #### What is your question? for one same audio(the audio file is long, such as...
RT 请问如何在pipeline中使用外挂的transformer-lm做shallow fusion
## 🐛 Bug punc入参的文本为空时报错如下: ` Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.cuda.DoubleTensor instead (while checking arguments for embedding)...