Commit 912a4bd5 authored by Jean Ibarz's avatar Jean Ibarz
Browse files

Add README.md

parent f0b6397a
# ML Binaural Audio
## Goals
The aim of this project is to apply Machine Learning techniques to binaural audio sound source localization. This project may lead to numerous works and scientific contributions:
* the proposal of novel neural networks architecture that may achieve or outperform state of art algorithms,
* generation of novel results, which may lead to an increase knowledge in terms of sound source localization performance,
* etc.
## Usage
Install dependancies:
> pip install -r requirements.txt
Run one of the scripts located in ./scripts : see next section for more details.
## Repository structure
| Project path | Description |
| :------ |:-----------|
| ./core | Package containing the core code. |
| ./databases/ | Folder containing data of multiple databases (IRCAM, CIPIC, etc.), stored as Pandas.Dataframe objects in pickle binary files. |
| ./scripts | Some useful python scripts. |
| ./scripts/generate_database.py | Generate files stored in ./databases from RAW databses files, usually raw .wav files or .mat files. |
| ./scripts/train2.py | Train a model. |
| ./scripts/generate_results2.py | Generate results, stored as .csv files, from one or multiple trained models stored in a root directory. |
| ./r-scripts | Generate some plots from stored results .csv files. |
# Some links that you may find interesting
Some binaural localization models: https://docs.google.com/document/d/1H5lV5_VNZ3JGffw8hrTtF9f8u89AS0pDubqrTOAHs3U/edit?usp=sharing
\ No newline at end of file
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment