SUPERB: Speech processing Universal PERformance Benchmark

Self-supervised learning (SSL) has proven vital for advancing research in natural language processing (NLP) and computer vision (CV). The paradigm pretrains a shared model on large volumes of unlabeled data and achieves state-of-the-art (SOTA) for various tasks with minimal adaptation. However, the speech processing community lacks a similar setup to systematically explore the paradigm. To bridge this gap, we introduce Speech processing Universal PERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the performance of a shared model across a wide range of speech processing tasks with minimal architecture changes and labeled data. Among multiple usages of the shared model, we especially focus on extracting the representation learned from SSL for its preferable re-usability. We present a simple framework to solve SUPERB tasks by learning task-specialized lightweight prediction heads on top of the frozen shared model. Our results demonstrate that the framework is promising as SSL representations show competitive generalizability and accessibility across SUPERB tasks. We release SUPERB as a challenge with a leaderboard(1) and a benchmark toolkit(2) to fuel the research in representation learning and general speech processing.

Automated summary

Self-supervised learning is important in advancing research in NLP and CV, but there is a lack of a similar setup for speech processing. The SUPERB benchmark is introduced to evaluate the performance of a shared model across various speech tasks, with a focus on utilizing representations learned from SSL. Results show promising generalizability and accessibility of SSL representations across SUPERB tasks.