STFT Generator and Visualizer

Project for CS 6221 @ GWU

This project provides a GUI-based tool to generate the Short-Time Fourier Transform (STFT) for a given WAV audio file. It also supports visualization and basic audio processing functionalities, making it a comprehensive tool for signal analysis and exploration.

---

Team Members

• Alice Dragnea
• Taslim Mahbub
• Akshay Nayak
• Jacob Weber

---

Features

• STFT Generation: Generate and visualize the spectrogram for WAV audio files.
• Inverse STFT: Reconstruct audio from the STFT.
• Line Chart Comparison: Visualize the original and reconstructed signals for quality assessment.
• GUI Interface: A web-based interface for easy interaction.
• Standalone Processing: Perform audio operations without the GUI.

---

Running the Project

1. Launch the Web Server:

   • Run the following command:

     go run .

   • This will launch the Gin web server.

2. Access the GUI:

   • Open your browser and navigate to:

     http://localhost:8080/
     

---

Code Structure

• Main Functionality:

  • The GUI and server logic are in main.go.
  • Navigate to http://localhost:8080/ to interact with the GUI.

• Audio Processing:

  • The audio processing logic resides in ssp.go.
  • Key functions:
    • perform_stft_pipeline:
      • Generates and saves the spectrogram.
      • Performs inverse STFT.
      • Saves line charts and reconstructed audio files.
    • perform_stft_standalone:
      • Performs only the STFT operations without additional processing.

---

External Dependencies

1. GoSSP:

   • Ensure you have the GoSSP library installed: https://github.com/r9y9/gossp/tree/master

2. Gnuplot:

   • Install Gnuplot for line chart visualization: http://www.gnuplot.info/

---

Branches

1. main:

   • Contains the GUI web server and the integrated STFT processing code.

2. mfcc-experiments:

   • Extends the audio processing pipeline by adding MFCC functionality on top of STFT.

3. jacob-STFT:

   • An earlier branch focused on STFT development (now fully integrated into main).

---

To-Do List

1. Extend the pipeline to include:

   • MFCC generation.
   • Inverse MFCC for audio reconstruction.

2. Improve file compatibility:

   • Support for non-header WAV files.
   • Process other audio file formats.

3. Enhance functionality:

   • Correctly handle stereo tracks.

---

License

This project is shared for educational purposes and is part of CS 6221 coursework at GWU. Contributions and feedback are welcome.

---

Screenshots

---