how to extract pitch speech features in Kaldi use python

[luweishuang]

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?

[luweishuang]

another, I follow steps in "https://zhuanlan.zhihu.com/p/55371926" and use a same wav as input , but to find the fbank_feat = logfbank(sig, nfilt=80, lowfreq=20, dither=0, wintype='povey') doesn't same as the kaldi's scipt "compute-kaldi-pitch-feats" running result. Is any wrong in parameters?