MiniSUPERB: Lightweight Benchmark for Self-supervised Speech Models
This provides a more efficient evaluation tool for researchers and practitioners in speech processing, though it is incremental as it builds directly on SUPERB.
The paper tackles the high computational cost of the SUPERB benchmark for evaluating self-supervised speech models by introducing MiniSUPERB, a lightweight alternative that achieves Spearman's rank correlations of 0.954 and 0.982 with SUPERB while reducing computational cost by 97% in terms of MACs.
SUPERB was proposed to evaluate the generalizability of self-supervised learning (SSL) speech models across various tasks. However, it incurs high computational costs due to the large datasets and diverse tasks. In this paper, we introduce MiniSUPERB, a lightweight benchmark that efficiently evaluates SSL speech models with comparable results to SUPERB but lower computational costs significantly. We carefully select representative tasks, sample datasets, and extract model representations offline. Our approach achieves a Spearman's rank correlation of 0.954 and 0.982 with SUPERB Paper and SUPERB Challenge, respectively. Additionally, we reduce the computational cost by 97% in terms of Multiply-ACcumulate operations (MACs). Furthermore, we evaluate SSL speech models in few-shot scenarios and observe significant variations in their performance. To our knowledge, this is the first study to examine both the computational cost of the model itself and the cost of evaluating it on a benchmark.