arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2206.12693cs.CLcs.SDeess.AS
收藏

TEVR:利用Token熵方差降低改进语音识别

TEVR: Improving Speech Recognition by Token Entropy Variance Reduction

hajo UG(hajo有限责任公司);Institute for Neuro- and Bioinformatics, University of Lübeck(吕贝克大学神经与生物信息学研究所)

Hajo Nils Krabbenhöft, Erhardt Barth

摘要:

本文提出了TEVR,一个语音识别模型,旨在最大限度地减少变化的令牌熵w.r.t.到语言模型。这利用了这样一个事实,即如果语言模型无论如何都能可靠而准确地预测一个标记,那么声学模型在识别它时就不需要准确。我们用9亿个参数训练德语ASR模型,结果表明,在CommonVoice德语中,TEVR的单词出错率为3.64%,其通过字出错率的相对16.89%的降低而优于最好的报告结果。我们希望向社区发布我们经过充分训练的语音识别管道将在未来带来保护隐私的离线虚拟助手。

英文摘要:

This paper presents TEVR, a speech recognition model designed to minimize the variation in token entropy w.r.t. to the language model. This takes advantage of the fact that if the language model will reliably and accurately predict a token anyway, then the acoustic model doesn't need to be accurate in recognizing it. We train German ASR models with 900 million parameters and show that on CommonVoice German, TEVR scores a very competitive 3.64% word error rate, which outperforms the best reported results by a relative 16.89% reduction in word error rate. We hope that releasing our fully trained speech recognition pipeline to the community will lead to privacy-preserving offline virtual assistants in the future.

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。

↑