Resource-constrained stereo singing voice cancellation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914647851925504 |
|---|---|
| author | Borrelli, Clara Rae, James Basaran, Dogac McVicar, Matt Souden, Mehrez Mauch, Matthias |
| author_facet | Borrelli, Clara Rae, James Basaran, Dogac McVicar, Matt Souden, Mehrez Mauch, Matthias |
| contents | We study the problem of stereo singing voice cancellation, a subtask of music source separation, whose goal is to estimate an instrumental background from a stereo mix. We explore how to achieve performance similar to large state-of-the-art source separation networks starting from a small, efficient model for real-time speech separation. Such a model is useful when memory and compute are limited and singing voice processing has to run with limited look-ahead. In practice, this is realised by adapting an existing mono model to handle stereo input. Improvements in quality are obtained by tuning model parameters and expanding the training set. Moreover, we highlight the benefits a stereo model brings by introducing a new metric which detects attenuation inconsistencies between channels. Our approach is evaluated using objective offline metrics and a large-scale MUSHRA trial, confirming the effectiveness of our techniques in stringent listening tests. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_12068 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Resource-constrained stereo singing voice cancellation Borrelli, Clara Rae, James Basaran, Dogac McVicar, Matt Souden, Mehrez Mauch, Matthias Sound Machine Learning Audio and Speech Processing We study the problem of stereo singing voice cancellation, a subtask of music source separation, whose goal is to estimate an instrumental background from a stereo mix. We explore how to achieve performance similar to large state-of-the-art source separation networks starting from a small, efficient model for real-time speech separation. Such a model is useful when memory and compute are limited and singing voice processing has to run with limited look-ahead. In practice, this is realised by adapting an existing mono model to handle stereo input. Improvements in quality are obtained by tuning model parameters and expanding the training set. Moreover, we highlight the benefits a stereo model brings by introducing a new metric which detects attenuation inconsistencies between channels. Our approach is evaluated using objective offline metrics and a large-scale MUSHRA trial, confirming the effectiveness of our techniques in stringent listening tests. |
| title | Resource-constrained stereo singing voice cancellation |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2401.12068 |