Audio-sentiment-analysis

This is a project for the Machine Learning course in Roma Tre University.

The datasets used are available at

• https://www.kaggle.com/ejlok1/toronto-emotional-speech-set-tess
• https://www.kaggle.com/uwrfkaggler/ravdess-emotional-speech-audio

Usage

To run the project

Download this repo or clone it by running

git clone https://github.com/luigibvl/Audio-sentiment-analysis

Move inside the downloaded directory

cd Audio-sentiment-analysis-master

You can record your own audio track to detect sentiment on by running

python3 record.py

to run the project run

python3 main.py