Simple Speech Features

Introduction

simple-speech-features is a simple, straight-forward Python demonstration for extracting and manipulating acoustic and prosodic features from sound waves based on Praat and Parselmouth. All speech extraction utilities and helper functions are included in feature_extraction_utils.py. The default values for parameters and information in the docstrings are directly from Praat.

Getting Started

1. Clone this repository: git clone https://github.com/uzaymacar/speech-features.git
2. Install Parselmouth: pip install praat-parselmouth
3. Start extracting speech features on your datasets as shown in main.ipynb!

References

• Jadoul, Y., Thompson, B., & de Boer, B. (2018). Introducing Parselmouth: A Python interface to Praat. Journal of Phonetics, 71, 1-15. https://doi.org/10.1016/j.wocn.2018.07.001
• http://www.fon.hum.uva.nl/praat/manual/Query.html
• http://www.fon.hum.uva.nl/rob/NKI_TEVA/TEVA/HTML/Analysis.html
• Feinberg, D. R. (2019, October 8). Parselmouth Praat Scripts in Python. https://doi.org/10.17605/OSF.IO/6DWR3
• http://practicalcryptography.com/miscellaneous/machine-learning/guide-mel-frequency-cepstral-coefficients-mfccs/
• Sample sound file (sample.wav) acquired from http://www.voiptroubleshooter.com/open_speech/american.html