Algorithm used to detect if an input .wav file is backward or forward. Algorithm can read and analyze a single .wav file or a folder containing more .wav files. Algorithm can also analyze multi-channels files: it analyzes each channel separately.
datafolder, containing some .wav files to make the jupyter notebooks runnotebooksfolder, containing some notebooks with files analysis and a "Recap" notebook, containing recap of the methodology adopted and the ideas used developing the code.srcfolder, containing the python code.gitignoreChangelog.mdREADME.mduv.lockpyproject.toml
Install dependencies with uv sync or make install.
To use the algorithm, run it using
python -m reverse_detection [path]
placing in place of [path] the path of a single .wav file or of a folder containing more .wav file for batch processing.
- Use
python -m reverse_detection -hto print the help message with the arguments required. - Use
python -m reverse_detection -f [path]if you want to apply a, high-pass filter, in the preprocessing phase, that 'stops' from 0 to 800Hz. This filter helps with classical, country, metal, rock music and, in some cases, also with signals containing voice recordings.
Others flags are available: check help message to read the complete list.
- Fabio Zanini
- Matteo Spanio
- Alessandro Russo
- Sergio Canazza
- University of Padova
If you use this code in your research, please cite the following paper:
@article{bosi2024a,
author = {Bosi, Marina and Zanini, Fabio and Spanio, Matteo and Russo, Alessandro and Sergio, Canazza},
journal = {Journal of the Audio Engineering Society},
title = {A novel derivative-based approach for the automatic detection of time-reversed audio in the MPAI/IEEE-CAE ARP international standard},
year = {2024},
number = {10190},
url = {https://aes2.org/publications/elibrary-page/?id=22693}
}This project is licensed under the GPL-3.0 License - see the LICENSE file for details.