This project focuses on emotion detection through sound analysis using machine learning techniques. The system is capable of discerning various emotions from audio clips, making it valuable for applications like sentiment analysis, voice-based assistants, and more.
- Utilizes a comprehensive dataset containing diverse audio samples associated with specific emotions.
- Performs exploratory data analysis (EDA) to ensure a balanced dataset.
- Extracts essential features from audio files using librosa, such as Mel-frequency cepstral coefficients (MFCCs).
- Builds a robust LSTM (Long Short-Term Memory) model using TensorFlow and Keras.
- Implements dropout layers for regularization to prevent overfitting.
- Achieves impressive results in emotion classification through meticulous training and evaluation.
- Provides a user-friendly interface for testing the model with custom audio files.
- Technology stack: Python, TensorFlow, Keras, Librosa, Seaborn.
Contributions are welcome! Please fork the repository and submit a pull request.
This project is licensed under the MIT License.
- Special thanks to Librosa and TensorFlow for their invaluable contributions to this project.
- The dataset used in this project is sourced from Kaggle.
For any inquiries or suggestions, please contact @sarthak.