-
This issue is in part discussed in https://github.com/librosa/librosa/issues/595, however, my question is more narrow.
It is quite common in speech processing to have win_length different from n_ff…
-
## [jit] aten::linspace error on Python 3.6
For torchaudio's CI setup (https://github.com/pytorch/audio/blob/master/build_tools/travis/install.sh), pytorch nightly is used on linux. The job seems to …
-
Hi - I tried your alternate model, and it worked good easily, so I am thankful for your work.
But I noticed the output of your melspectrogram() function clips to 1.0 often on LJSpeech data.
(Of cou…
-
Does anybody have such a problem? When it is trained for 1000k steps with LjSpeech , the "abrupt noise" appears. For example:
![image](https://user-images.githubusercontent.com/40649244/50745732-bbe7…
-
health@health-desktop:~/Desktop/lang_detec/Speech-to-Text-WaveNet-master$ python recognize.py --file test.wav
/usr/local/lib/python2.7/dist-packages/numba/errors.py:104: UserWarning: Insufficiently r…
-
Hi,
I am trying to run a 1D Conv on a Melspectrogram with kapre, but it seems like the Melspectrogram layer assumes that 2D operation will be done subsequently by giving a 4D output.
So at this mom…
-
The enclosure is my test with fixed learning rate of 1e-4. We can see the evaluation deteriorate from 275k to 300k. It seemed the learning failed to converge.
[fixed_lr.zip](https://github.com/fatcho…
-
Format of node_js/src/weights/RES8_NARROW.js and the weight file created by training.py do not have same format. How to convert the file created by training.py into format similar to node_js/src/weig…
-
When I try to use the notebook to generate spectrogram for training a vocoder, I get the following results as spectrogram (plz note it's upside down):
![Screen Shot 2019-09-09 at 12 56 10 PM](https…
-
```
...
File "/u/zeyer/setups/librispeech/2018-02-26--att/returnn/TFEngine.py", line 1180, in train
line: self.train_epoch()
locals:
self =
self.train_epoch =
File "…