Menglong Xu
Menglong Xu
I also meet this error. To solve it, I just check the length before carry out "aws_last_success[0, :, 0].nonzero()[:, -1].max()". If it got 0, I set rightmost_frame to 0. But...
you can add "causal: True" (or False) in the config file like this: 
Thank you for your question. We found that one parameter was missing in a previous code change, which may be the reason for the model's non-convergence. In wekws/kws/utils/executor.py, line 48:...
We have checked the result on xiaowen dataset using ds_tcn model. the result seems good now ( after the aforementioned bug fixed ).
We have added cache in mdtc, please see https://github.com/wenet-e2e/wekws/pull/105.
fixed, please see https://github.com/wenet-e2e/wekws/pull/105.
Thanks, you can infer the approximate end time-stamp of the keyword from the scores in scores.txt.
> ``` > pip install chardet > pip install cchardet > ``` > > if you switch to later pytorch versions. Worked for me on pytorch 1.13 Thanks, it also...
For streaming wakeup word detection, you can follow this guideline (https://github.com/wenet-e2e/wekws/blob/main/runtime/android/README.md). If you need assistance in training your model, you can refer to the "run.sh" script for a step-by-step guide...
@taorui-plus 你可以试下设置 num_layers=4，应该能解决导出的问题。 这个 error 是因为 mdtc 和 tcn 中表示层数的变量名称不一致导致的，tcn 里的 num_layers 实际上对应于 mdtc 中的 num_stack