The Speech Emotion Recognition Music App is an intelligent application that detects human emotions from speech input and plays suitable music using SoundCloud based on the recognized emotion.
The goal of this project is to enhance user experience by combining speech processing, machine learning, and emotion-aware music recommendation.
- Speech input through microphone or audio files
- Emotion detection from speech signals
- Emotion classification (Happy, Sad, Angry, Neutral, Fearful, Disgust, Surprise)
- Automatic music playback from SoundCloud based on detected emotion
- User-friendly and interactive interface
| Emotion | Music Type Played |
|---|---|
| Happy | Upbeat, energetic tracks |
| Sad | Calm, soothing tracks |
| Angry | Relaxing, soft tracks |
| Neutral | Light, ambient music |
| Fear | Comforting, slow music |
| Disgust | Clean, mellow, mood-lifting tracks |
| Surprise | Energetic, exciting, dynamic tracks |
- Python
- Machine Learning / Deep Learning
- Librosa – audio feature extraction
- NumPy, Pandas – data processing
- Scikit-learn / TensorFlow / PyTorch – model training
- SoundCloud – music streaming
- Streamlit – web interface
- Capture speech input
- Preprocess audio signal
- Extract speech features (MFCC, Chroma, Spectral features, etc.)
- Predict emotion using trained ML model
- Play emotion-based music in SoundCloud
- Select your favourite artist from the drop-down and record duration (in seconds) from the right pane.
- Click on "Click to Record" and record your audio.
⚠️ Note: Make sure your microphone is enabled and working.- While recording, the microphone emoji is red.
- After recording is complete, the microphone emoji turns blue.
- Click on "Analyze Emotion" to detect your emotions and get song recommendations.
- Pick a song from the recommended list and click on "Play on SoundCloud" to listen.
- Harineesha Nutakki
- Attili Valli Sai Meghana