-
I'm getting this error after running "bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package "
**System information**
- Windows 10
- TensorFlow version: 1.13
- Python version:…
-
i found any feature the kaldi performed will pass function "build_LFR_features", is it has any special function? what's means by "stacking frames and skipping frames"?
-
From kaldi-help, ["Purpose and use of utt2dur"](https://groups.google.com/forum/#!topic/kaldi-help/vyEL7qtXmhk):
@kkm000:
> I always carry [the `utt2dur`] file, compute it carefully to the fifth d…
-
I would like to apologize in advance if this is the wrong place to ask a question. I have a question regarding the data normalization and the pca feature creation. If your dataset had a shape of (batc…
-
I see that `features/normalization.py` implements a couple of useful normalisations/standardisations but I'm unsure where in the code it would be best to add them. Any thoughts?
-
Managed to do mfcc and vad - checked data to be valid.
Now when train_diag_ubm is ran I get error job failed without further information
Utterances are relative long, it's 90 people each recorded fo…
-
I would like to use my own features whiche were saved in files. So, I have inspected features "inFeat " in the Featurize.cpp code and I have removed these three lines:
auto& mfcc = getMfcc();
featS…
-
Hi Mr. Dogan,
In the past few days, I was trying to create a GMM/HMM model using pyKaldi and I found the script named `gmm-recognizer.py` in the `examples` directory. And I came across the part whe…
-
Hi,
Is there a way to convert the MFCC features that extracted with Kaldi (.ark file) to a .csv file that include all the parameters of the MFCC matrix?
Thanks
-
The following code extracts the time/fequencydomain features for the given signal
mfccs = (librosa.feature.mfcc(y=X, sr=sample_rate, n_mfcc=20).T)
print("np.shape(mfcc)", np.shape(mfccs))
chroma …