基于粗标签的高效CTC正则化端到端语音翻译
Efficient CTC Regularization via Coarse Labels for End-to-End Speech Translation
摘要：
对于端到端的语音翻译，使用源转录本或目标翻译作为标签，使用连接时间分类（CTC）目标正则化编码器可以大大提高质量度量。然而，CTC需要在词汇空间上额外的预测层，带来不可忽略的模型参数和计算开销，尽管该层通常不用于推理。在本文中，我们重新审视了CTC正则化对真正词汇标签的需求，并探索了减少CTC标签空间的策略，目标是在不降低质量的情况下提高效率。我们提出了粗标记CTC（CoLaCTC），通过简单的启发式规则，如使用截断，除法或模（MOD）操作合并词汇标签。尽管它很简单，但我们在4种源语言和8种目标语言上的实验表明，CoLaCTC特别是MOD可以将标签空间压缩到256甚至更高，获得训练效率（1.18倍~1.77倍加速，取决于原始词汇量），但仍然提供与CTC基线相当或更好的性能。我们还表明，CoLaCTC成功地推广到CTC正则化，无论使用转录或翻译标记。
英文摘要：
For end-to-end speech translation, regularizing the encoder with the Connectionist Temporal Classification (CTC) objective using the source transcript or target translation as labels can greatly improve quality metrics. However, CTC demands an extra prediction layer over the vocabulary space, bringing in nonnegligible model parameters and computational overheads, although this layer is typically not used for inference. In this paper, we re-examine the need for genuine vocabulary labels for CTC for regularization and explore strategies to reduce the CTC label space, targeting improved efficiency without quality degradation. We propose coarse labeling for CTC (CoLaCTC), which merges vocabulary labels via simple heuristic rules, such as using truncation, division or modulo (MOD) operations. Despite its simplicity, our experiments on 4 source and 8 target languages show that CoLaCTC with MOD particularly can compress the label space aggressively to 256 and even further, gaining training efficiency (1.18x ~ 1.77x speedup depending on the original vocabulary size) yet still delivering comparable or better performance than the CTC baseline. We also show that CoLaCTC successfully generalizes to CTC regularization regardless of using transcript or translation for labeling.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
