This project contains the implementation of the following preprocessing and features of a signal:
- Importing speech file and sampling
- Preprocessing
- Frame blocking
- Windowing
- Preemphasis
- Extract features
- Time-domain features
- Mean
- Energy
- ZCR (Zero Crossing Rate)
- Auto-correlation
- Normal implementaion
- Implementing by Multiplying the signal to a power of an odd number
- Implementing with Center Clipping
- Implementing with 3-level Center Clipping
- AMDF (Average Magnitude Difference Function)
- Frequency-domain features
- FFT (Fast Fourier Transform)
- LPC (Linear predictive coding)
- Cepstral Analysis
- Spectrum Estimation
- Spectrum Estimation based on FFT
- Spectrum Estimation based on LPC
- Pitch Estimation
- Pitch Estimation based on Auto Correlation
- Time-domain features
Also, the used libraries in this project are as follows:
librosa
: a python package for music and audio analysisscipy
: a free and open-source Python library used for scientific computing and technical computingmatplotlib
: a comprehensive library for creating static, animated, and interactive visualizations in Pythonnumpy
: a library for the Python programming language, adding support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays