arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.07178eess.AScs.AIcs.CLcs.LGcs.SD
收藏

一种用于端到端语音识别的拼写校正模型

A spelling correction model for end-to-end speech recognition

Google NYC(谷歌纽约)

Jinxi Guo, Tara N. Sainath, Ron J. Weiss

摘要:

用于语音识别的基于注意力的序列到序列模型使用单个神经网络联合训练声学模型、语言模型(LM)和对齐机制,并且仅需要并行的音频-文本对。因此,端到端模型的语言模型组件仅在转录的音频-文本对上进行训练,这导致性能下降,特别是在罕见单词上。虽然已经有各种各样的工作着眼于将在纯文本数据上训练的外部LM整合到端到端框架中,但它们都没有考虑到模型的特征误差分布。在本文中,我们提出了一种新的方法来利用纯文本数据,通过训练拼写校正(SC)模型来明确纠正这些错误。在LibriSpeech数据集上,我们证明了该模型在直接纠正顶级ASR假设时,WER比基线模型相对提高了18.6%,在使用外部LM进一步重新评分扩展的n-最佳列表时,WER相对提高了29.0%。

英文摘要:

Attention-based sequence-to-sequence models for speech recognition jointly train an acoustic model, language model (LM), and alignment mechanism using a single neural network and require only parallel audio-text pairs. Thus, the language model component of the end-to-end model is only trained on transcribed audio-text pairs, which leads to performance degradation especially on rare words. While there have been a variety of work that look at incorporating an external LM trained on text-only data into the end-to-end framework, none of them have taken into account the characteristic error distribution made by the model. In this paper, we propose a novel approach to utilizing text-only data, by training a spelling correction (SC) model to explicitly correct those errors. On the LibriSpeech dataset, we demonstrate that the proposed model results in an 18.6% relative improvement in WER over the baseline model when directly correcting top ASR hypothesis, and a 29.0% relative improvement when further rescoring an expanded n-best list using an external LM.

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。

↑