This repository contains Jupyter Notebook–based experiments for basic audio forensics / audio analysis tasks developed as a university project. The work was done as part of coursework and represents my early attempts at applying Python and audio libraries to forensic-style problems.
Demonstrates exploratory approaches to audio forensics tasks, such as: - Loading and visualizing audio signals - Basic spectral and time‑domain analysis - Feature extraction (e.g., MFCCs, spectrograms) - Simple classification or comparison experiments Focus was on learning and experimentation rather than polished tools.
This was an early university assignment I completed when I was still learning how to handle projects, so: - Code may be messy, duplicated, or poorly modularized. - Notebooks may contain one-off cells, incomplete explanations, and ad-hoc data handling. - Reproducibility is not guaranteed — paths, hardcoded variables, and missing requirements may exist. - There are likely missing tests, limited error handling, and no CI.
At the time I did this project I was still learning how to: - Structure a repository for reproducibility. - Use version control effectively across branches and features. - Write modular, testable code instead of ad-hoc notebook cells. - Document assumptions, data sources, and environment requirements.