2020/05/08 by Manuel Pariente, Samuele Cornell, Pariente, Manuel +25 · 7 citations
Computer Science · #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Music and Audio Processing #Sound (cs.SD) #Speech Recognition and Synthesis #Speech and Audio Processing #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.04132
openalex publication_date 2020/05/08 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
This paper describes Asteroid, the PyTorch-based audio source separation\ntoolkit for researchers. Inspired by the most successful neural source\nseparation systems, it provides all neural building blocks required to build\nsuch a system. To improve reproducibility, Kaldi-style recipes on common audio\nsource separation datasets are also provided. This paper describes the software\narchitecture of Asteroid and its most important features. By showing\nexperimental results obtained with Asteroid's recipes, we show that our\nimplementations are at least on par with most results reported in reference\npapers. The toolkit is publicly available at\nhttps://github.com/mpariente/asteroid .\n