利用 LLM 增强进行上下文 ASR 错误处理，以服务于目标导向对话式 AI
Contextual ASR Error Handling with LLMs Augmentation for Goal-Oriented Conversational AI
AI总结：
针对目标导向对话中缺乏先验数据且语言多变的 ASR 错误，提出 LLM 上下文增强与双向排序纠错方法，在真实家庭装修和烹饪场景中显著提升召回率与 F1。
摘要：
通用自动语音识别（ASR）系统在目标导向对话中并不总是表现良好。现有 ASR 纠错方法依赖于先前的用户数据或命名实体。我们将纠错扩展到没有先前用户数据且表现出语言灵活性（如词汇和句法变化）的任务。我们提出了一种新颖的大语言模型上下文增强方法和排序策略，该策略融合来自目标导向对话式 AI 对话状态及其任务的上下文信息。我们的方法对两类对象进行排序：（1）依据 n-best ASR 假设与上下文的词汇和语义相似性对其排序；（2）依据上下文与 ASR 假设的语音对应关系对上下文排序。在家庭装修和烹饪领域通过真实用户进行评估，我们的方法将纠错的召回率和 F1 分别提高了 34% 和 16%，同时保持了精确率和假阳性率。当纠错方法正常工作时，用户评分提高了 0.8 至 1 分（满分 5 分），且没有因假阳性而导致评分下降。
英文摘要：
General-purpose automatic speech recognition (ASR) systems do not always perform well in goal-oriented dialogue. Existing ASR correction methods rely on prior user data or named entities. We extend correction to tasks that have no prior user data and exhibit linguistic flexibility such as lexical and syntactic variations. We propose a novel context augmentation with a large language model and a ranking strategy that incorporates contextual information from the dialogue states of a goal-oriented conversational AI and its tasks. Our method ranks (1) n-best ASR hypotheses by their lexical and semantic similarity with context and (2) context by phonetic correspondence with ASR hypotheses. Evaluated in home improvement and cooking domains with real-world users, our method improves recall and F1 of correction by 34% and 16%, respectively, while maintaining precision and false positive rate. Users rated .8-1 point (out of 5) higher when our correction method worked properly, with no decrease due to false positives.
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。
