通过在线优化伪标签，提升语音识别模型性能
Enhancing BEST-RQ Pseudo-Label Quality through Online Refinement for Automatic Speech Recognition
- 用PCA替代线性投影，改进伪标签生成
- 迭代更新码本，使伪标签更准确
- 适合追求简单高效自监督语音模型的研究者
BEST-RQ是一种简洁有效的自监督语音表征学习方法，在自动语音识别（ASR）任务中表现良好。它采用固定在线量化方案生成伪标签，虽简化了训练流程，但提供的监督信号弱于HuBERT类模型的迭代优化方式。本文提出三项改进：将量化器的线性投影替换为主成分分析（PCA），通过迭代码本精炼更新码本，并引入基于码本蒸馏的额外码本。在LibriSpeech 960小时数据集上预训练，使用100小时有监督数据微调。三项改进全部启用后，LibriSpeech测试集（test-other）的词错误率（WER）相对降低12%，从10.1%降至8.8%。
原文摘要 · Abstract (English)
BEST-RQ is a simple and effective self-supervised training method for speech representation learning that performs well on automatic speech recognition (ASR) tasks. It generates pseudolabels using a fixed online quantization scheme, which simplifies training but provides weaker supervision than HuBERT-style models that iteratively refine pseudo-labels. In this work, we improve online pseudo-label generation while preserving simplicity. We propose three modifications: replacing the quantizer's linear projection with Principal Component Analysis (PCA), updating the codebook via iterative codebook refinement, and introducing an additional codebook updated via codebook distillation. We pre-train on the LibriSpeech 960-hour dataset and fine-tune using 100 hours of supervised LibriSpeech data. With all three modifications enabled, we achieve a 12% relative reduction in word error rate (WER) on the LibriSpeech test-other set, improving from 10.1% to 8.8%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。