Generalizing audio source separation
with large-scale data (GASS)


arXiv


Universal source separation targets at separating the audio sources of an arbitrary mix, removing the constraint to operate on a specific domain like speech or music. Yet, the potential of universal source separation is limited because most existing works focus on mixes with predominantly sound events, and small training datasets also limit its potential for supervised learning. Here, we study a single general audio source separation (GASS) model trained to separate speech, music, and sound events in a supervised fashion with a large-scale dataset. We assess GASS models on a diverse set of tasks. Our strong in-distribution results show the feasibility of GASS models, and the competitive out-of-distribution performance in sound event and speech separation shows its generalization abilities. Yet, it is challenging for GASS models to generalize for separating out-of-distribution cinematic and music content. We also fine-tune GASS models on each dataset and consistently outperform the ones without pre-training. All fine-tuned models (except the music separation one) obtain state-of-the-art results in their respective benchmarks.


Upstream

Listen how our models work in-distribution

Downstream

Listen how our models work out-of-distribution


Select which kind of data (upstream task) you want to listen:



Example 1: speech + guitar + vehicle

Mixture Source 1 Source 2 Source 3 Source 4
no source #4
(silence)
TDANet-Wav
TDANet-STFT
BSRNN

Example 2: speech + speech

Mixture Source 1 Source 2 Source 3 Source 4
no source #3
(silence)
no source #4
(silence)
TDANet-Wav
TDANet-STFT
BSRNN