RNN to classify accents by using YouTube Biritsh and American training samples
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Sound-Data
data
.gitattributes
README.md
main.py

README.md

Sound Classification

This recurrent neural network tries to differentiate between sounds. The RNN is trained by using example data provided in the repo. in Sound-Data folder.


Installation

> - Clone Repository
> - Install Dependencies

Training

> - python main.py -m train

Testing

> - python main.py -m pred -u <model ID from trained model here>

Dependiencies

tensorflow
numpy
librosa
matplotlib
tqdm
uuid

Example

The RNN successfully classificated 1.wav as label 0 in class dictionary img_pred