Modern automatic speech recognition (ASR) models, such as OpenAI's Whisper, rely on deep encoder-decoder architectures, and their encoders are a critical bottleneck for efficient deployment due to high computational intensity. We introduce LiteASR, a low-rank compression scheme for ASR encoders that significantly reduces inference costs while maintaining transcription accuracy. Our approach leverages the strong low-rank properties observed in intermediate activations: by applying principal component analysis (PCA) with a small calibration dataset, we approximate linear transformations with a chain of low-rank matrix multiplications, and further optimize self-attention to work in reduced dimensionality. Evaluation results show that our method can compress Whisper large-v3's encoder size by over 50%, matching Whisper medium's size with better transcription accuracy, thereby establishing a new Pareto frontier of accuracy and efficiency. The code of LiteASR is available at https://github.com/efeslab/LiteASR.
@article{arxiv.2502.20583,
title = {LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation},
author = {Keisuke Kamahori and Jungo Kasai and Noriyuki Kojima and Baris Kasikci},
journal= {arXiv preprint arXiv:2502.20583},
year = {2025}
}