Contextual Speech Recognition with Difficult Negative Training Examples
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)(2018)
摘要
Improving the representation of contextual information is key to unlocking the potential of end-to-end (E2E) automatic speech recognition (ASR). In this work, we present a novel and simple approach for training an ASR context mechanism with difficult negative examples. The main idea is to focus on proper nouns (e.g., unique entities such as names of people and places) in the reference transcript, and use phonetically similar phrases as negative examples, encouraging the neural model to learn more discriminative representations. We apply our approach to an end-to-end contextual ASR model that jointly learns to transcribe and select the correct context items, and show that our proposed method gives up to $53.1\%$ relative improvement in word error rate (WER) across several benchmarks.
更多查看译文
关键词
speech recognition,sequence-to-sequence models,phonetics,attention,biasing
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络