# Resource-Efficient Separation Transformer

Source: [Resource-Efficient Separation Transformer](<https://www.sota2.com/research/paper/68fa71ca1e209131cea52fd5>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:12:32.787Z

Authors: Luca Della Libera, Cem Subakan, Mirco Ravanelli, Samuele Cornell, Fr\\'ed\\'eric Lepoutre, Fran\\c{c}ois Grondin

Published: 2022-06-19T23:37:24.000Z

## Abstract

Transformers have recently achieved state-of-the-art performance in speech separation. These models, however, are computationally demanding and require a lot of learnable parameters. This paper explores Transformer-based speech separation with a reduced computational cost. Our main contribution is the development of the Resource-Efficient Separation Transformer (RE-SepFormer), a self-attention-based architecture that reduces the computational burden in two ways. First, it uses non-overlapping blocks in the latent space. Second, it operates on compact latent summaries calculated from each chunk. The RE-SepFormer reaches a competitive performance on the popular WSJ0-2Mix and WHAM! datasets in both causal and non-causal settings. Remarkably, it scales significantly better than the previous Transformer-based architectures in terms of memory and inference time, making it more suitable for processing long mixtures.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Multi-target Sound Extraction on FSD Kaggle + TAU Urban Acoustic Scenes synthetic mixture 2018 2019 (test)](<https://www.sota2.com/research/sota/multi-target-sound-extraction-on-fsd-kaggle-plus-tau-urban-acoustic-scenes-synthetic-mixture-2018-2019-test.md>): Multi-target Sound Extraction; FSD Kaggle + TAU Urban Acoustic Scenes synthetic mixture 2018 2019 (test); SI-SNRi (1 Class): 7.42.

- [Single-target Sound Extraction on FSD Kaggle 2018 and TAU Urban Acoustic Scenes 2019 (test)](<https://www.sota2.com/research/sota/single-target-sound-extraction-on-fsd-kaggle-2018-and-tau-urban-acoustic-scenes-2019-test.md>): Single-target sound extraction; FSD Kaggle 2018 and TAU Urban Acoustic Scenes 2019 (test); SI-SNRi: 7.26.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2206.09507>)

[PDF](<https://arxiv.org/pdf/2206.09507.pdf>)