Guardado en:
Detalles Bibliográficos
Autores principales: Chen, Junyu, Vekkot, Susmitha, Shukla, Pancham
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2309.08684
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910372392337408
author Chen, Junyu
Vekkot, Susmitha
Shukla, Pancham
author_facet Chen, Junyu
Vekkot, Susmitha
Shukla, Pancham
contents Music source separation (MSS) aims to extract 'vocals', 'drums', 'bass' and 'other' tracks from a piece of mixed music. While deep learning methods have shown impressive results, there is a trend toward larger models. In our paper, we introduce a novel and lightweight architecture called DTTNet, which is based on Dual-Path Module and Time-Frequency Convolutions Time-Distributed Fully-connected UNet (TFC-TDF UNet). DTTNet achieves 10.12 dB cSDR on 'vocals' compared to 10.01 dB reported for Bandsplit RNN (BSRNN) but with 86.7% fewer parameters. We also assess pattern-specific performance and model generalization for intricate audio patterns.
format Preprint
id arxiv_https___arxiv_org_abs_2309_08684
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Music Source Separation Based on a Lightweight Deep Learning Framework (DTTNET: DUAL-PATH TFC-TDF UNET)
Chen, Junyu
Vekkot, Susmitha
Shukla, Pancham
Audio and Speech Processing
Sound
Music source separation (MSS) aims to extract 'vocals', 'drums', 'bass' and 'other' tracks from a piece of mixed music. While deep learning methods have shown impressive results, there is a trend toward larger models. In our paper, we introduce a novel and lightweight architecture called DTTNet, which is based on Dual-Path Module and Time-Frequency Convolutions Time-Distributed Fully-connected UNet (TFC-TDF UNet). DTTNet achieves 10.12 dB cSDR on 'vocals' compared to 10.01 dB reported for Bandsplit RNN (BSRNN) but with 86.7% fewer parameters. We also assess pattern-specific performance and model generalization for intricate audio patterns.
title Music Source Separation Based on a Lightweight Deep Learning Framework (DTTNET: DUAL-PATH TFC-TDF UNET)
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2309.08684