This program processes any given .wav signal (or records a signal via input) and creates its respective LPC and MFCC vectors, with configurable params. It then calculates the similarity between inputs through DTW.
This project is largely based on the code/information provided by the following resources and their respective authors:
- Linear Predictive Coding, kunigami, 11 Dec 2021
- LPC_Speech_Synthesis
- StackOverflow: "Writing WAV file using Python, Numpy array and WAVE module"
- vanilla-stft-mfcc
This is a command-line Python program. Python 3 is necessary to run. It is also possible to run this program via the given Dockerfile (through Makefile), however, you'll need to modify the Makefile to include a display port for Docker in order to see Pyplot graphs. It's recommended to run this program directly on your computer without Docker.
To run this program on your local computer, execute the following Git command:
git clone https://github.com/andrejanesic/Python-Word-Similarity-Analysis.git
This will clone the repository onto your local computer.
This program requires Python, so if it isn't installed on your computer, you can download it from the Python.org website.
Next, install the required libraries for Python using pip:
pip install -r requirements.txt
You're good to go!
This program was developed as an individual assignment for the class of [5025] Speech Recognition, cohort 2022/23, at the School of Computing, Union University, Belgrade.
#TODO