Jointly Detecting and Separating Singing Voice: A Multi-Task Approach
This work addresses the problem of singing voice separation and detection for music processing applications, but it is incremental as it builds on existing multi-task learning methods with specific adaptations.
The paper tackles the challenge of limited training data in music processing by proposing a multi-task approach that jointly learns vocal activity detection and vocal separation, resulting in improved performance for both tasks compared to single-task baselines.
A main challenge in applying deep learning to music processing is the availability of training data. One potential solution is Multi-task Learning, in which the model also learns to solve related auxiliary tasks on additional datasets to exploit their correlation. While intuitive in principle, it can be challenging to identify related tasks and construct the model to optimally share information between tasks. In this paper, we explore vocal activity detection as an additional task to stabilise and improve the performance of vocal separation. Further, we identify problematic biases specific to each dataset that could limit the generalisation capability of separation and detection models, to which our proposed approach is robust. Experiments show improved performance in separation as well as vocal detection compared to single-task baselines. However, we find that the commonly used Signal-to-Distortion Ratio (SDR) metrics did not capture the improvement on non-vocal sections, indicating the need for improved evaluation methodologies.