This repository contains some personal projects of mine related to audio and DSP.
Me and my colleague, Rahul Peter participated in the Query by Vocal Imitation Challenge and our model ranked at third place in the challenge. The abstract of the report was presented in the Late Breaking section during the AES AIMLA Conference, 2025, London. More details about this project can be found in this repository.
This folder contains Faust implementations of some gain, modulation and time-based effects; the idea is to try and incorporate some of these effects as unit generators as part of Arco, a signal processing server created by Dr. Roger Dannenberg. You can check out the Arco repository here.
The signals under consideration in numerous engineering applications, including voice synthesis, analysis, radar, sonar, and telecommunications, are known to be nonstationary, meaning that their properties change over time. A typical method for power spectral density (PSD) estimate cannot be used for spectral analysis of these kinds of data. To cope with such signals, various techniques have been offered, including the time-frequency analysis.
For time-frequency analysis, one of the first techniques was the short-time Fourier transform (STFT). A portion of the signal is removed by a moving window, and the local characteristics of the signal are obtained from the Fourier transform of this portion. Nonstationary signals are analysed using the spectrogram, which is the squared magnitude of the STFT.
Another commonly used time-frequency distribution (TFD) is the Wigner–Ville distribution (WVD)
- The Wigner-Ville Distribution (WVD) is a mathematical tool used in signal processing to analyze non-stationary signals. It's named after Eugene Wigner and is a representation of a signal in the time-frequency domain. In essence, WVD provides a joint time-frequency representation of a signal, which means it shows how the frequency content of the signal changes over time. This is particularly useful for signals whose frequency components change over time, such as those encountered in communication systems, radar, sonar, and biomedical signal processing.
- Fourier-Bessel coefficients are coefficients used in expressing a function in terms of a series of orthogonal functions called Bessel functions. These coefficients play a role similar to the Fourier coefficients in expressing a function in terms of sine and cosine functions but for functions defined on a finite interval or a bounded domain, especially in cylindrical or spherical geometries.
The research project aims to develop a novel method for decomposing digital signals into time-frequency representations using a combination of the Wigner-Ville Distribution and Fourier-Bessel Coefficients, with the aim of reducing cross terms.