SongkhareSentiment-Aware Music Recommendation
An LSTM reads the mood in your words and picks the music to match.
Recommendation system that infers emotional context from text with an LSTM and maps predicted emotion onto music valence scores.
What I built
- LSTM model used to infer emotional context from textual input
- User text processed and predicted emotions mapped to music valence scores for personalised recommendations
- Full-stack inference workflow connecting the trained model to a web interface
- Recommendations generated in real time from user-provided mood and emotional context
Stack
- Python
- TensorFlow
- Keras
- LSTM
- NumPy
- Pandas