akisaini / dl-project Goto Github PK

README

EMOTION CLASSIFICATION USING SPEECH INPUT ON RAVDESS DATASET

Instructions for setup:

1. Download the dataset from the following link: https://zenodo.org/record/1188976#.YFZuJ0j7SL8

2. Make sure to choose the Audio Speech Dataset and not the Audio Song Dataset.

3. Once downloaded, extract and move folder to your working directory.

4. From the Code folder, download the file 'main_cnn.py', 'main_lstm.py' and 'main_rnn.py'.

5. Make sure the files are located in your current working directory. Run the files 'main_cnn.py' for generating a CNN model, 'main_lstm.py' for generating the LSTM model and 'main_rnn.py' for generating the RNN model.

6. The above files can be run using the command: python3 'file_name.py'. For example: python3 main_cnn.py

7. The py scripts use generalized path and path seperators thorughout the code and should work across platforms.