多语言上下文适配器以改进低资源语言中的自定义词识别
Multilingual Contextual Adapters To Improve Custom Word Recognition In Low-resource Languages
AI总结：
针对低资源语言自定义词识别问题，提出监督损失与多语言策略改进Contextual Adapters训练，实现48%的F1提升并降低基础CTC模型5-11%的WER。
摘要：
联结时序分类（CTC）模型因其在自动语音识别（ASR）中速度与性能的平衡而广受欢迎。然而，这些CTC模型在其他方面仍有困难，例如针对自定义词的个性化。最近的一种方法探索了Contextual Adapters，其中使用基于注意力的偏置模型用于CTC，以改进对自定义实体的识别。虽然该方法在数据充足时效果良好，但我们展示它对低资源语言并非有效策略。在本工作中，我们提出一种监督损失，用于更平滑地训练Contextual Adapters。此外，我们探索一种多语言策略，以在有限训练数据下提升性能。我们的方法在为低资源语言检索未见自定义实体时实现了48%的F1提升。有趣的是，作为训练Contextual Adapters的副产品，我们也看到基础CTC模型的性能实现了5-11%的词错误率（WER）降低。
英文摘要：
Connectionist Temporal Classification (CTC) models are popular for their balance between speed and performance for Automatic Speech Recognition (ASR). However, these CTC models still struggle in other areas, such as personalization towards custom words. A recent approach explores Contextual Adapters, wherein an attention-based biasing model for CTC is used to improve the recognition of custom entities. While this approach works well with enough data, we showcase that it isn't an effective strategy for low-resource languages. In this work, we propose a supervision loss for smoother training of the Contextual Adapters. Further, we explore a multilingual strategy to improve performance with limited training data. Our method achieves 48% F1 improvement in retrieving unseen custom entities for a low-resource language. Interestingly, as a by-product of training the Contextual Adapters, we see a 5-11% Word Error Rate (WER) reduction in the performance of the base CTC model as well.
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。
