Papers › Continual self-training with bootstrapped remixing for speech enhancement
Continual self-training with bootstrapped remixing for speech enhancement
Efthymios Tzinis, Yossi Adi, Vamsi K. Ithapu, Buye Xu, Anurag Kumar
We propose RemixIT, a simple and novel self-supervised training method for speech enhancement. The proposed method is based on a continuously self-training scheme that overcomes limitations from previous studies including assumptions for the in-domain noise distribution and having access to clean target signals. Specifically, a separation teacher model is pre-trained on an out-of-domain dataset and is used to infer estimated target signals for a batch of in-domain mixtures. Next, we bootstrap the mixing process by generating artificial mixtures using permuted estimated clean and noise signals. Finally, the student model is trained using the permuted estimated sources as targets while we periodically update teacher's weights using the latest student model. Our experiments show that RemixIT outperforms several previous state-of-the-art self-supervised methods under multiple speech enhancement tasks. Additionally, RemixIT provides a seamless alternative for semi-supervised and unsupervised domain adaptation for speech enhancement tasks, while being general enough to be applied to any separation task and paired with any separation model.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Speech Enhancement | Deep Noise Suppression (DNS) Challenge | Sudo rm-rf (U=8) | PESQ-WB | 2.69 | #26 of 36 | Archive leaderboard | report |
| Speech Enhancement | Deep Noise Suppression (DNS) Challenge | Sudo rm-rf (U=8) | SI-SDR-WB | 18.6 | #26 of 36 | Archive leaderboard | report |
| Speech Enhancement | Deep Noise Suppression (DNS) Challenge | RemixIT (w Sudo U=32) | PESQ-WB | 2.60 | #28 of 36 | Archive leaderboard | report |
| Speech Enhancement | Deep Noise Suppression (DNS) Challenge | RemixIT (w Sudo U=32) | SI-SDR-WB | 18.0 | #28 of 36 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections