-
### 🚀 The feature, motivation and pitch
# Preface
Hello! I am trying to export the codes for audio preprocessing on `kaldi` module in torchaudio.
I have few feature requests:
- Is there any …
-
Iunno if this is well known but i have github.com/judyfong/sid
and github.com/cadia-lvl/kaldi-speaker-diarization they are submodules of kaldi for the ruv-di dataset
-
```
In file included from /usr/ports/audio/kaldi/work/kaldi-4a8b7f6/src/matrix/compressed-matrix.cc:22:
In file included from /usr/ports/audio/kaldi/work/kaldi-4a8b7f6/src/matrix/../matrix/compresse…
-
Hi,
I would like to understand the differences between Whisper's fbank features and Kaldi's fbank features? I get that conceptually both features are derived from Mel-filterbank energies, but then wh…
-
I followed the very old instructions to build Kaldi for WASM here: https://gitlab.inria.fr/multispeech/kaldi.web/kaldi-wasm/-/wikis/build_details.md
I found out that the performance with this was v…
msqr1 updated
1 month ago
-
When setting up kaldi from a git clone it uses openfst 1.8.3
I also had to run the following to update opengrm tools as well as the specified version doesen't play well with c17 vs c11
```sed -i s…
-
With the openfst changes now merged upstream (https://github.com/kaldi-asr/kaldi/pull/4927), we could get rid of a lot of patches here.
https://github.com/conda-forge/kaldi-feedstock/blob/4183636f2c9…
-
The kaldi-models-0.03.zip in the install_models.sh script is no longer available because the website certificate has expired. Is there a new version of the model archive?
-
-
Hi there,
I am wondering what does fbank really give us in the dataloader? I went to torchaudio doc and did not find much info about what it is. Does anyone have a link to its explanation?
Thank…
lix4 updated
6 months ago