Assignment1: Real-Time FFT Window

This is a real-time FFT window for audio signal processing. It delivers a graphical interface for real-time audio signal processing, including denoising using a pre-trained model. The GUI is built using PyQt5 and pyqtgraph, and it relies on external libraries such as numpy, torch, soundfile and PyAudio for audio processing.

Features

• Real-time audio signal processing
• Denoising using a pre-trained model
• Graphical interface for plotting audio signals
• Adding artificial white Gaussian noise to the audio signal
• Saving input and output audio files
• Playing (saved) input and output audio files

Usage

1. Install the required dependencies by running pip install -r requirements.txt.
2. Position yourself at the root directory
3. Start the application from the root directory
   • python assignment1-signals\main_gui.py
   • additional command line arguments:
     • --dns48 or --dns64 or --master64 or --valentini_nc
       • different pre-trained models
     • --dry <float> between 0 and 1
       • level of noise removal, where 0 is maximum noise removal, but it can cause distortion. 0.04 is default value
4. Connect to the desired input and output audio devices.
5. Click the "Save" button to save the input and output audio files.
6. Click the "Listen to Audio" to load "input.wav" and "output.wav" files.
   • input.wav contains original signal recorded with the microphone + the AWGN
   • output.wav contains denoised signal
7. In the asignment1-signals/logs folder you can find the .log file for each run of the application

Dependencies

• python 3.8.18
• see requirements.txt

NOTE

• https://github.com/facebookresearch/denoiser/tree/main/denoiser

Resources

@inproceedings{defossez2020real, title={Real Time Speech Enhancement in the Waveform Domain}, author={Defossez, Alexandre and Synnaeve, Gabriel and Adi, Yossi}, booktitle={Interspeech}, year={2020} }

Assignment2: PCA 3D Morphing

This project is a software integration assignment that focuses on 3D PCA (Principal Component Analysis) for analyzing and manipulating 3D models. The main file, assignment2.py, serves as the entry point and incorporates several libraries, including PyQt5, OpenGL, numpy, scipy, and imageio.

Functionality

The project utilizes PCA to analyze and manipulate 3D models, offering users the ability to adjust texture and geometry weights. Notably, it includes multithreading functionality for efficient texture and geometry processing. The code showcases the integration of various libraries and provides a user-friendly interface for parameter adjustments. Users can manipulate weights for eigenvectors of the PCA using sliders in the separate parameters window.

Dependencies

• python 3.8.18
• see requirements.txt

Usage

1. Clone the directory.
   • git clone https://github.com/karlic-luka/Software_Integration_1.git
2. Install the required dependencies by running pip install -r requirements.txt.
3. Position yourself at the root directory
4. Start the application from the root directory
   • python assignment2_3D_PCA\assignment2.py
   • otherwise, there could be some problems with paths
5. Load the model1.obj by clicking on Load paths
6. Click Process to process PCA on texture and geometry
7. Play around with the sliders (SPOILER ALERT: geometry sliders are really fun!)
   • I challenge you to make as funny face as possible
8. When you're done, you can save the new model by clicking on Save 3D Face