/SSP_Project

Speech Emotion Detection using Prosody Features

Primary LanguageJupyter Notebook

SSP_Project

Speech Emotion Detection using Prosody Features

Project done as part of the Speech Signal Processing (SSP) course of the Monsoon 2020 semester at IIIT-H.

Teammate: Hasir Mushtaq

Instructions

  1. Copy the EmoDB database from here and extract it inside a folder named 'DB' inside the main directory.
  2. Run the following command to restructure the database:

    python arrangeFiles.py

  3. Run the following command to extract prosody features and save:

    bash FeatureExtraction/extractFeatures.sh

  4. Features will be saved in FeatureExtraction/features
  5. Use these features to run MLP.py or SSP.ipynb as per your setup to generate results.