对比两种新词识别方法，发现上下文偏置更稳定，语音大模型在读音场景表现好但易受干扰。
How to Recognize New Words: A Comparison Between Context Biasing Methods and Speech LLMs
- 用词表注入方式让ASR模型在推理时优先识别新词
- 上下文偏置使特定词汇错误率降低88%，对其他词影响小
- 语音大模型适合读音文本，但对干扰词和提示顺序敏感
识别新词与罕见词（如命名实体、缩写、领域专有词等）仍是自动语音识别（ASR）的核心挑战。本文比较了两种策略：一种是上下文偏置方法，即扩展ASR模型使其在推理时可输入词表；另一种是直接以上下文提示语音大语言模型（Speech LLM）。我们在朗读与非朗读语音上评估了基于Whisper的两种上下文偏置方法与三种语音LLM，报告了偏置词、非偏置词及整体词错误率（WER）。结果显示，上下文偏置方法将偏置词的WER降低了高达88%相对值，且对其他词影响甚微；语音LLM在朗读语音上表现优异，但在非朗读语音上泛化能力较弱，且对干扰词数量和提示词顺序敏感。本文分析了两者的权衡关系，为方法选择提供依据。
原文摘要 · Abstract (English)
Recognizing new and rare words - named entities, acronyms, domain specific special words, and other items scarce in training data - remains a key challenge for automatic speech recognition (ASR). We compare two strategies for this: context biasing methods, where an ASR model is extended such that during inference a word list can be supplied, and speech large language models (LLMs) prompted with context directly. We evaluate two context biasing methods based on Whisper against three speech LLMs across read and non-read speech, reporting biased, unbiased, and overall word error rate (WER). The context biasing methods cut biased WER by up to 88% relative while leaving other words largely unaffected. Speech LLMs excel on read speech but generalize less well to non-read speech, and prove sensitive to distractor count and prompt word order. We characterize the resulting trade-offs to guide method selection.
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。