Multiple Choice Learning for Efficient Speech Separation with Many Speakers
This work addresses a core challenge in speech separation for audio processing applications, offering a more efficient alternative to existing methods.
The paper tackles the permutation problem in supervised speech separation by replacing Permutation Invariant Training (PIT) with Multiple Choice Learning (MCL), achieving comparable performance on WSJ0-mix and LibriMix benchmarks while being computationally more efficient.
Training speech separation models in the supervised setting raises a permutation problem: finding the best assignation between the model predictions and the ground truth separated signals. This inherently ambiguous task is customarily solved using Permutation Invariant Training (PIT). In this article, we instead consider using the Multiple Choice Learning (MCL) framework, which was originally introduced to tackle ambiguous tasks. We demonstrate experimentally on the popular WSJ0-mix and LibriMix benchmarks that MCL matches the performances of PIT, while being computationally advantageous. This opens the door to a promising research direction, as MCL can be naturally extended to handle a variable number of speakers, or to tackle speech separation in the unsupervised setting.