https://github.com/adityaapraveen/speech-emotion-recognition

This project implements a deep learning model for recognizing emotions in speech using Long Short-Term Memory (LSTM) networks. It processes speech signals to extract features, trains an LSTM-based neural network, and accurately predicts the emotional state of the speaker.
https://github.com/adityaapraveen/speech-emotion-recognition

Last synced: about 13 hours ago
JSON representation

This project implements a deep learning model for recognizing emotions in speech using Long Short-Term Memory (LSTM) networks. It processes speech signals to extract features, trains an LSTM-based neural network, and accurately predicts the emotional state of the speaker.

• Host: GitHub
• URL: https://github.com/adityaapraveen/speech-emotion-recognition
• Owner: adityaapraveen
• Created: 2024-03-21T13:05:11.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-03-21T13:15:54.000Z (8 months ago)
• Last Synced: 2024-04-25T05:28:47.050Z (7 months ago)
• Language: Jupyter Notebook
• Size: 81.1 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Speech Emotion Recognition with LSTM

This repository contains a deep learning model for Speech Emotion Recognition (SER) using Long Short-Term Memory (LSTM) networks. SER is the task of automatically recognizing the emotional state of a person from their speech signals. The model is trained on a dataset of speech recordings labeled with corresponding emotions. It extracts Mel-Frequency Cepstral Coefficients (MFCCs) as features from the speech signals and feeds them into an LSTM-based neural network for classification. The trained model can accurately predict the emotional state (e.g., happiness, sadness, anger) of a speaker based on their speech.

## Key Features

- Utilizes MFCC features extracted from speech signals for emotion recognition.

- Implements an LSTM-based neural network for sequence modeling and classification.

- Provides training, evaluation, and prediction functionalities for speech emotion recognition tasks.

- Includes visualization tools such as confusion matrix and classification report for performance analysis.

## Usage

1. **Preprocess Your Dataset**: Preprocess your dataset and extract MFCC features from speech recordings.

2. **Train the Model**: Train the LSTM model using the provided code or load a pre-trained model.

3. **Evaluate Model Performance**: Evaluate model performance using classification reports, confusion matrices, and ROC curves.

4. **Predict Emotions**: Use the trained model for predicting emotions in new speech recordings.

## Dependencies

- Python 

- NumPy

- pandas

- scikit-learn

- TensorFlow or PyTorch (for LSTM implementation)

- librosa

- matplotlib

- seaborn

## Contributing

Contributions to this project are welcome! Please feel free to open issues or submit pull requests with any improvements, bug fixes, or additional features.