让语音模型提前看未来，精准识别生僻人名地名。
Peeking Into The Future For Contextual Biasing
- 用候选实体列表引导模型同时预测多个未来词，实现上下文偏置。
- 在LibriSpeech上，罕见实体错误率降低50.34%。
- 无需额外模块，简化结构，适合虚拟助手等实用场景。
端到端自动语音识别（ASR）模型虽在通用转录中表现优异，但在识别罕见或未见过的命名实体（如联系人姓名、地点）方面仍存在困难，而这对虚拟助手等下游应用至关重要。本文提出一种基于注意力机制的编码器-解码器（AED）模型的上下文偏置方法，利用候选命名实体列表进行增强。该方法不局限于预测下一个词，而是同时预测多个未来词，使模型能够“窥探未来”，对候选实体列表中的潜在实体进行评分。此外，本方法直接使用多词预测的逻辑值，无需额外实体编码器或交叉注意力层，显著降低架构复杂度。在LibriSpeech数据集上的实验表明，相比基线AED模型，该方法在命名实体词错误率上实现了最高50.34%的相对提升。
原文摘要 · Abstract (English)
While end-to-end (E2E) automatic speech recognition (ASR) models excel at general transcription, they struggle to recognize rare or unseen named entities (e.g., contact names, locations), which are critical for downstream applications like virtual assistants. In this paper, we propose a contextual biasing method for attention based encoder decoder (AED) models using a list of candidate named entities. Instead of predicting only the next token, we simultaneously predict multiple future tokens, enabling the model to "peek into the future" and score potential candidate entities in the entity list. Moreover, our approach leverages the multi-token prediction logits directly without requiring additional entity encoders or cross-attention layers, significantly reducing architectural complexity. Experiments on Librispeech demonstrate that our approach achieves up to 50.34% relative improvement in named entity word error rate compared to the baseline AED model.
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。