TEVR: Improving Speech Recognition by Token Entropy Variance Reduction

Hajo Nils Krabbenhöft, Erhardt Barth


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.
Original languageEnglish
Number of pages10
Specialist publicationarXiv.org
Publication statusPublished - 25.06.2022

Research Areas and Centers

  • Centers: Center for Artificial Intelligence Luebeck (ZKIL)
  • Research Area: Intelligent Systems


Dive into the research topics of 'TEVR: Improving Speech Recognition by Token Entropy Variance Reduction'. Together they form a unique fingerprint.

Cite this