Speech Self-Supervised Representation Benchmarking: Are We Doing it Right?
This work highlights a critical flaw in current benchmarking practices for speech SSL, potentially affecting researchers and practitioners who rely on these evaluations for model development.
The paper investigates the robustness of self-supervised speech representation benchmarks to changes in decoder architecture, finding that varying decoders leads to significant variations in leaderboards and may cause counterproductive increases in SSL model sizes.
Self-supervised learning (SSL) has recently allowed leveraging large datasets of unlabeled speech signals to reach impressive performance on speech tasks using only small amounts of annotated data. The high number of proposed approaches fostered the need and rise of extended benchmarks that evaluate their performance on a set of downstream tasks exploring various aspects of the speech signal. However, and while the number of considered tasks has been growing, most rely upon a single decoding architecture that maps the frozen SSL representations to the downstream labels. This work investigates the robustness of such benchmarking results to changes in the decoder architecture. Interestingly, it appears that varying the architecture of the downstream decoder leads to significant variations in the leaderboards of most tasks. Concerningly, our study reveals that benchmarking using limited decoders may cause a counterproductive increase in the sizes of the developed SSL models.