# How to extract prosodic cues from a wav file using Python

How do I extract the following features as from wav files in Python -

1. Voicing rate (Voice rate) - frequency of phonemes while speaking, and represents the pace of a conversation
2. Number of autocorrelation peaks
3. Pitch, pitch bandwidth

This project Speech Signal Processing Toolkit (SPTK) provides several features you are looking for. Here is a good wrapper around it https://github.com/r9y9/pysptk This wrapper is using a slightly different version of SPTK, but installation is straightforward.

pip install pysptk

1) I would use kaldi or another speech recognition package on phones output and a VAD system to compute the voicing rate.

2) For autocorrelation, convert the waveform to mel-cepstrum with the function mcep then compute the autocorrelation with c2acr

3) Here is an example to compute the Pitch http://nbviewer.jupyter.org/github/r9y9/pysptk/blob/master/examples/pysptk%20introduction.ipynb