This program leverages the powerful audio feature extraction capabilities of Praat, enhanced with configurations made by Shahabks. The primary goal of this tool is to extract detailed audio features from datasets such as:
- syllables mean
- average syllable duration
- F2 range
- speech rate mean
- pauses mean
- speaking duration mean
- pause duration mean
- original duration mean
- speaking balance mean
- speaking duration percentage
- pausing duration percentage
- articulation rate mean
- F0 mean mean
- local jitter average
- local absolute jitter average
- local shimmer average
- local db shimmer average
- mean harmonics to noise - mean HNR
- F1 average
- F2 average
- F3 average
- F4 average
- amplitude difference H1-A3
- pitch average
- mean intensity
It is highly recommended to install conda to avoid potential issues with missing C libraries.
- Clone an existing base environment:
conda create --name new-env-name --clone base
- Clone repository
git clone https://github.com/Matic-M/Audio-feature-extraction
- Install parselmouth
pip install praat-parselmouth
- Ensure build tools and dependencies are installed
pip install -r requirements.txt
- if building wheels fails, try prebuilt wheels
- For more information visit readthedocs.io and maintained Parselmouth