Instant One-Shot Word-Learning for Context-Specific Neural Sequence-to-Sequence Speech Recognition

Abstract

Neural sequence-to-sequence systems deliver state-of-the-art performance for automatic speech recognition (ASR). When using appropriate modeling units, e.g., byte-pair encoded characters, these systems are in principal open vocabulary systems. In practice, however, they often fail to recognize words not seen during training, e.g., named entities, numbers or technical terms. To alleviate this problem we supplement an end-to-end ASR system with a word/phrase memory and a mechanism to access this memory to recognize the words and phrases correctly. After the training of the ASR system, and when it has already been deployed, a relevant word can be added or subtracted instantly without the need for further training. In this paper we demonstrate that through this mechanism our system is able to recognize more than 85% of newly added words that it previously failed to recognize compared to a strong baseline.

Recent Advances in End-to-End Automatic Speech Recognition

“…In this way, the network output conditions not only on the speech signal and previous label sequence but also on the contextual phrase list. It was further extended into a two-step memory enhanced model in [190]. It is common that a contextual phrase list contains rare words, especially names which are not observed during training.…”