Real-time Low-latency Music Source Separation using Hybrid Spectrogram-TasNet
This addresses the need for efficient music source separation in applications like hearing aids and live shows, representing an incremental advance.
The paper tackles the problem of adapting music demixing models for real-time low-latency applications, proposing HS-TasNet which achieves an SDR of 4.65 at 23 ms latency on MusDB, improving to 5.55 with more data.
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.