Sure! Here’s the markdown version of your project description without the example code:

Music Mood Recognizer 🎶💡

This project predicts the mood of a song based on its audio features, such as danceability, energy, valence, tempo, speechiness, and acousticness. It uses machine learning to classify the mood into predefined categories and provides mood-based music recommendations by playing YouTube playlists.

Key Features ✨

• Mood Prediction: Input song features (e.g., danceability, energy) and predict the mood of the song.
• Music Recommendations: Based on the predicted mood, get a mood-specific playlist that is automatically played via YouTube.
• Real-Time Playlists: Integrates with pywhatkit to play mood-based music playlists on YouTube.
• User-Friendly Interface: Simple, interactive tool to input song features and get recommendations.

Tech Stack 🛠️

• Machine Learning:
  • Classification algorithms (Random Forest)
  • Trained using the Spotify Tracks Dataset
• Python:
  • Data preprocessing with Pandas & NumPy
  • Model training using Scikit-Learn
• YouTube Integration:
  • Play music via YouTube using pywhatkit
• Deployment: End-to-end solution for mood prediction and song recommendations ; used Flask.

Setup Instructions 🚀

1. Clone this repository:

   git clone https://github.com/AmandaHanz/music-mood-recognizer.git

2. Install the necessary Python libraries

3. Download the model and dataset:

   • Download the Spotify Tracks Dataset from Kaggle and place it in the project folder.
   • Train the model or use a pre-trained model provided in the repository.

4. Input song features when prompted and get mood-based music recommendations!

Mood-Based Playlist Recommendations 🎧

• Happy songs playlist
• Sad songs playlist
• Energetic workout music
• Calm relaxing tunes
• Romantic love songs
• Upbeat dance hits
• Mysterious suspenseful tracks
• Intense cinematic soundtracks
• Peaceful nature sounds
• Darkwave synth music

Model Development 💻

• Data Collection & Preprocessing: Utilized the Spotify Tracks Dataset from Kaggle to extract key audio features and preprocess the data for model training.
• Model Development: Applied classification algorithms (such as Random Forest) to predict mood labels based on the extracted features. Trained the model using a labeled dataset with various mood categories.
• Mood Prediction: Developed an interactive tool where users can input song features (such as danceability, energy, etc.) and receive a mood prediction for the song.
• Music Recommendation: Integrated the system with YouTube to play mood-based song playlists using pywhatkit library, enabling users to listen to recommended songs in real time.
• Deployment: Built an end-to-end solution that allows users to input song features, predicts the mood, and provides a direct music link based on the prediction.