zitengwang / python_kaldi_features Goto Github PK

Hi, could you share your fbank array value and the one that Kaldi computes of the wav file? Thanks.

I found in task asr, pitch feature in kaldi is much more popular. It compute fbank feature and pitch feature and I want to extract this pitch feature in python, can you give some useful ways to do that?

Hi，
Thanks for your sharing, It helps me so much

I got almost exactly same results by use logfbank(sig2,nfilt=23,lowfreq=20,dither=0,wintype='povey') with compute-fbank-feats --dither=0.0 in Kaldi
But when I change something like dither, nfilt, I got different features when I do in kaldi,
The closest I can get is with dither=0.0

So I think maybe something was different in DCT or lifter ?
Then I checked dct and lifter,
For dct, you use the DCT(II), Kaldi use DCT(II) too.
For lifter, the equation of lifter in your is same as Kaldi.
But the decimal part is still different

Could you please tell what makes the difference or I do something wrong?
Is it an error in calculation?
Thank you : )

Kaldi use DCT(II) :
The following is ComputeDctMatrix from Kaldi's ComputeDctMatrix which I think is different from dct of yours:
`template void ComputeDctMatrix(Matrix *M) {
//KALDI_ASSERT(M->NumRows() == M->NumCols());
MatrixIndexT K = M->NumRows();
MatrixIndexT N = M->NumCols();

KALDI_ASSERT(K > 0);
KALDI_ASSERT(N > 0);
Real normalizer = std::sqrt(1.0 / static_cast(N)); // normalizer for
// X_0.
for (MatrixIndexT j = 0; j < N; j++) (*M)(0, j) = normalizer;
normalizer = std::sqrt(2.0 / static_cast(N)); // normalizer for other
// elements.
for (MatrixIndexT k = 1; k < K; k++)
for (MatrixIndexT n = 0; n < N; n++)
(*M)(k, n) = normalizer
* std::cos( static_cast(M_PI)/N * (n + 0.5) * k );
}`

Hello,

I am new to kaldi..

It is very hard to find a command-line which produces the same result from example.py.

Can you upload a sample command-line??