Here’s an updated version of your README to reflect the changes for using Gradio and running app.py:

This project classifies music genres using a Recurrent Neural Network (RNN) trained on audio features extracted from music files. It leverages the librosa library for audio processing, tensorflow for the model, and pydub for audio format conversion. The accuracy is 55% (for now).

• Installation
• Usage
• Dependencies
• How It Works
• Contributing
• License

Make sure you have Python installed on your machine. You can set up the required packages using pip. Create a virtual environment and install the dependencies as follows:

pip install pandas librosa tensorflow pydub numpy gradio

1. Run the Gradio Interface: You can start the Gradio interface to predict the genre of audio files.

   python Music_Genre_Classifier/app.py

2. Upload an Audio File: Open the Gradio interface in your web browser http://127.0.0.1:7860 (or the link provided in the console after running the script). Upload your .mp3 audio file through the interface.

3. View Predictions: The predicted genre will be displayed on the interface after processing the audio file.

• pandas: For data manipulation and analysis.
• librosa: For audio and music analysis.
• tensorflow: For building and loading the neural network model.
• pydub: For audio file format conversion.
• numpy: For numerical operations.
• gradio: For creating a user-friendly interface for model predictions.

1. Feature Extraction: The audio file is loaded, and Mel-frequency cepstral coefficients (MFCC) are extracted to represent the audio features.

2. Model Loading: The pre-trained RNN model on the GTZAN dataset is loaded from the specified file path.

3. Audio Conversion: If your input file is in .mp3 format, it is converted to .wav using pydub.

4. Prediction: The extracted features are fed into the model to predict the genre. The output is the genre label corresponding to the predicted index.

5. Output: The predicted genre is displayed in the Gradio interface.

Feel free to submit issues or pull requests if you would like to contribute to this project.

This project is licensed under the MIT License.

Feel free to modify any sections further based on your specific project needs or additional features!