A Multi-Task Learning Framework for Sound Event Detection using High-level Acoustic Characteristics of Sounds
Domestic environment sound event detection task leveraging some distinctive high-level acoustic characteristics of various sound events to assist the SED model training, without requiring additional labeled data.
Multitask Learning DCASE Task 4 recipe:
Challenge website here
Go to ./recipes/dcase2022_task4_baseline
and follow the instructions there in the README.md
In the recipe, we provide a conda script that creates a suitable conda environment with all dependencies, including pytorch with GPU support in order to run the recipe. There are also instructions for data download and preparation.
Run python setup.py install
to install the desed_task package
If this work is helpful, please feel free to cite the following paper:
"Tanmay Khandelwal and Rohan Kumar Das, “A Multi-Task Learning Framework for Sound Event Detection using High-level Acoustic Characteristics of Sounds”, in Proc. Interspeech 2023, Dublin, Ireland, August 2023."
To access the paper:
Arxiv