Attention-Driven Time-Domain Convolutional Network for Source Separation of Vocal and Accompaniment
Gorde:
| Argitaratua izan da: | Electronics vol. 14, no. 20 (2025), p. 3982-4009 |
|---|---|
| Egile nagusia: | |
| Beste egile batzuk: | , , , |
| Argitaratua: |
MDPI AG
|
| Gaiak: | |
| Sarrera elektronikoa: | Citation/Abstract Full Text + Graphics Full Text - PDF |
| Etiketak: |
Etiketarik gabe, Izan zaitez lehena erregistro honi etiketa jartzen!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3265895189 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 2079-9292 | ||
| 024 | 7 | |a 10.3390/electronics14203982 |2 doi | |
| 035 | |a 3265895189 | ||
| 045 | 2 | |b d20250101 |b d20251231 | |
| 084 | |a 231458 |2 nlm | ||
| 100 | 1 | |a Zhao, Zhili |u College of Computer Science and Technology, Qingdao University, Qingdao 266071, China; 2018204625@qdu.edu.cn (Z.Z.); sch_mail@163.com (C.S.); src@qdu.edu.cn (R.S.) | |
| 245 | 1 | |a Attention-Driven Time-Domain Convolutional Network for Source Separation of Vocal and Accompaniment | |
| 260 | |b MDPI AG |c 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a Time-domain signal models have been widely applied to single-channel music source separation tasks due to their ability to overcome the limitations of fixed spectral representations and phase information loss. However, the high acoustic similarity and synchronous temporal evolution between vocals and accompaniment make accurate separation challenging for existing time-domain models. These challenges are mainly reflected in two aspects: (1) the lack of a dynamic mechanism to evaluate the contribution of each source during feature fusion, and (2) difficulty in capturing fine-grained temporal details, often resulting in local artifacts in the output. To address these issues, we propose an attention-driven time-domain convolutional network for vocal and accompaniment source separation. Specifically, we design an embedding attention module to perform adaptive source weighting, enabling the network to emphasize components more relevant to the target mask during training. In addition, an efficient convolutional block attention module is developed to enhance local feature extraction. This module integrates an efficient channel attention mechanism based on one-dimensional convolution while preserving spatial attention, thereby improving the ability to learn discriminative features from the target audio. Comprehensive evaluations on public music datasets demonstrate the effectiveness of the proposed model and its significant improvements over existing approaches. | |
| 653 | |a Feature extraction | ||
| 653 | |a Musical instruments | ||
| 653 | |a Machine learning | ||
| 653 | |a Music | ||
| 653 | |a Deep learning | ||
| 653 | |a Musicians & conductors | ||
| 653 | |a Separation | ||
| 653 | |a Musical performances | ||
| 653 | |a Neural networks | ||
| 653 | |a Time domain analysis | ||
| 653 | |a Attention | ||
| 653 | |a Modules | ||
| 653 | |a Target masking | ||
| 653 | |a Acoustics | ||
| 653 | |a Singers | ||
| 653 | |a Information retrieval | ||
| 700 | 1 | |a Luo, Min |u Arts College, Qingdao University, Qingdao 266071, China | |
| 700 | 1 | |a Qiao Xiaoman |u Information Technology Department, Qingdao Vocational and Technical College of Hotel Management, Qingdao 266100, China; 2020020653@qdu.edu.cn | |
| 700 | 1 | |a Shao Changheng |u College of Computer Science and Technology, Qingdao University, Qingdao 266071, China; 2018204625@qdu.edu.cn (Z.Z.); sch_mail@163.com (C.S.); src@qdu.edu.cn (R.S.) | |
| 700 | 1 | |a Sun Rencheng |u College of Computer Science and Technology, Qingdao University, Qingdao 266071, China; 2018204625@qdu.edu.cn (Z.Z.); sch_mail@163.com (C.S.); src@qdu.edu.cn (R.S.) | |
| 773 | 0 | |t Electronics |g vol. 14, no. 20 (2025), p. 3982-4009 | |
| 786 | 0 | |d ProQuest |t Advanced Technologies & Aerospace Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3265895189/abstract/embedded/6A8EOT78XXH2IG52?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text + Graphics |u https://www.proquest.com/docview/3265895189/fulltextwithgraphics/embedded/6A8EOT78XXH2IG52?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3265895189/fulltextPDF/embedded/6A8EOT78XXH2IG52?source=fedsrch |