Semantic Scholar Open Access 2024 11 sitasi

Real-Time Low-Latency Music Source Separation Using Hybrid Spectrogram-Tasnet

Satvik Venkatesh Arthur Benilov Philip Coleman Frederic Roskam

Lihat Sumber DOI

Abstrak

There have been significant advances in deep learning for music demixing in recent years. However, there has been little attention given to how these neural networks can be adapted for real-time low-latency applications, which could be helpful for hearing aids, remixing audio streams and live shows. In this paper, we investigate the various challenges involved in adapting current demixing models in the literature for this use case. Subsequently, inspired by the Hybrid Demucs architecture, we propose the Hybrid Spectrogram Time-domain Audio Separation Network (HS-TasNet), which utilises the advantages of spectral and waveform domains. For a latency of 23 ms, the HS-TasNet obtains an overall signal-to-distortion ratio (SDR) of 4.65 on the MusDB test set, and increases to 5.55 with additional training data. These results demonstrate the potential of efficient demixing for real-time low-latency music applications.

Topik & Kata Kunci

Engineering Computer Science

Penulis (4)

Satvik Venkatesh

Arthur Benilov

Philip Coleman

Frederic Roskam

Format Sitasi

APA MLA BibTeX

Venkatesh, S., Benilov, A., Coleman, P., Roskam, F. (2024). Real-Time Low-Latency Music Source Separation Using Hybrid Spectrogram-Tasnet. https://doi.org/10.1109/ICASSP48485.2024.10448381

Akses Cepat

Lihat di Sumber doi.org/10.1109/ICASSP48485.2024.10448381

Informasi Jurnal

Tahun Terbit: 2024
Bahasa: en
Total Sitasi: 11×
Sumber Database: Semantic Scholar
DOI: 10.1109/ICASSP48485.2024.10448381
Akses: Open Access ✓