weedwind

How to return the individual loss of each sequence?

1

Hi, Right now the code returns the average or total loss for the entire batch, I am wondering how should I change the source code to return loss for each...

Does the code internally normalize the mean and the length of the ivectors?

1

Hi, RicherMans, Usually before we train the PLDA model, we subtract the mean from the ivectors and also normalize their length. I am wondering does your implementation (or Kaldi's) do...

Does it support the latest Kaldi?

3

Hi, RicherMans, I installed the latest Kaldi. When I make files, it reports /home/xliu/PLDA/src/kaldi-utils.hpp: In function ‘kaldi::Matrix kaldi::readFeatureFromPyString(PyObject*)’: /home/xliu/PLDA/src/kaldi-utils.hpp:44:41: error: ‘PyString_Check’ was not declared in this scope PyString_Check(pyfeaturefilename); ^ /home/xliu/PLDA/src/kaldi-utils.hpp:45:74:...

CPU RAM memory leak when using the beam search decoder

1

Hi, I used the ctc beam decoder from this link https://github.com/joshemorris/pytorch-ctc. However, I found that after I finished decoding one utterance, the decoder does not release RAM memory. After decoding...

Hi, I encountered an issue during installation. Does anyone know how to solve it? Preparing metadata (setup.py) ... error error: subprocess-exited-with-error python setup.py egg_info did not run successfully. exit code:...

Can you get the initial mean SDR on LibriSpeech using Google's test list?

8

Hi, seungwonpark, I was trying to use Google's posted [test list](https://github.com/google/speaker-id/tree/master/publications/VoiceFilter/dataset/LibriSpeech) for LibriSpeech to reproduce their results. But I can not even get their initial mean SDR (10.1 dB in...

Hi, 你好。我叫刘晓宇，我读了这篇paper和你的code，方便加个微信讨论一下吗？ 我的邮箱是 [email protected]，可以把你的微信号发给我吗？多谢多谢。

Can you open source the code for generating the synthetic echo data?

1

Hi, I am wondering if you are able to provide the code that you used to generate those synthetic echo data?