-
We have 5 (or more?) different binary formats for the tagger that are supported. At least:
* Unigram model 1
* Unigram model 2
* Unigram model 3
* 2-gram HMM
* Perceptron
And additionally t…
-
Have you tried building the spectrogram and encoder output in smaller chunks and appending? I think the spectrogram should generate fairly easily with minimal noise depending on the size of the chunk,…
-
### Describe the bug
Making this issue mostly for myself, and to track/document/point others to the problem.
The LibriSpeech conformer-transducer recipe fails to train with the stock parameters.…
-
When trying to write a DataFrame to an EDF file, with physical maximums/minimums in precisions over 16 bits, the text defaults to something like 'Physical minimum for channel xx...' even when it shoul…
-
torch 1.8.0+cu111
cuda 11.1
```
`(base) jovyan@liubiao-0:~/apex$ pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
/opt/cond…
-
python 3.7.11
nvcc: 11.1, V11.1.74
Build cuda_11.1.TC455_06.29069683_0
pytorch:1.8.0+cu111
cudnn:8005
RTX3090
(python37) ldf@ldf-virtual-machine:~/.jupyter/qinghao/PanGu/panguAlpha_pytorch/ape…
-
environment：
python=3.7
pytorch==1.0.1 torchvision==0.2.2 cudatoolkit=9.0
pillow==6.0.0
CUDA Version 9.0.176
error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1
Running set…
-
**Describe the bug**
After following installation instructions for Linux there is a weird bug about torch tensor not being on CPU when converting to numpy.
```
Transcribing: 0%| …
-
I'm seeking guidance on how to incorporate a Phone-Based Language Lexicon (in icefall/egs/librispeech/ASR/prepare.sh in Step 6) into the latest Zipformer Model, a state-of-the-art solution in speech r…
-
(Adding on issues board for documentation, this PR will be out over the week.)
Wu and Cotterell's papers on strong alignment seem just up our alley for the library. There should be an implementatio…