通过纠错反馈提升语音识别中拼写发音不一致词的识别准确率
Context Biasing for Pronunciation-Orthography Mismatch in Automatic Speech Recognition
- 利用推理时用户实时纠正错误，反向优化模型对生僻词的识别
- 在偏置词错误率上相对文本替换法提升22%~34%
- 适合需要高精度识别专有名词、缩略词的场景
序列到序列神经模型在自动语音识别中表现优异，使用字节对编码等建模单元时理论上具备开放词汇能力。然而实际中仍难以识别训练未见的词汇，如命名实体、缩略词或领域专用词。尽管已有多种上下文偏置方法，但当出现发音与拼写不一致时效果受限。本文提出一种新方法：利用替换错误的纠错信息，在推理阶段动态修正识别结果。用户可实时添加纠正内容。实验表明，该方法相较传统文本替换法，在偏置词错误率上实现22%至34%的相对提升，同时保持整体性能稳定。
原文摘要 · Abstract (English)
Neural sequence-to-sequence systems deliver state-of-the-art performance for automatic speech recognition. When using appropriate modeling units, e.g., byte-pair encoding, these systems are in principle open vocabulary systems. In practice, however, they often fail to recognize words not seen during training, e.g., named entities, acronyms, or domain-specific special words. To address this problem, many context biasing methods have been proposed; however, these methods may still struggle when they are unable to relate audio and corresponding text, e.g., in case of a pronunciation-orthography mismatch. We propose a method where corrections of substitution errors can be used to improve the recognition accuracy of such challenging words. Users can add corrections on the fly during inference. We show that with this method we get a relative improvement in biased word error rate between 22% and 34% compared to a text-based replacement method, while maintaining the overall performance.
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。