HackerPapers

arXiv:2206.12693cs.CL

TEVR: Improving Speech Recognition by Token Entropy Variance Reduction

Hajo Nils Krabbenhöft, Erhardt Barth

Abstract

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.

arXiv comments
10 pages including 2 pages appendix, 1 figure, 6 tables
Journal reference
DOI
Subjects
cs.CLComputation and Language cs.SDSound eess.ASAudio and Speech Processing

On Hacker News

Linked by 2 identifiable commenters in 4 comments across 4 threads, from Aug 2022 to Jun 2026.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. Finding Optimal Tokenizers 1 comment

    • This is an interesting approach with integer programming and then using an explicit solver. It’s probably very slow, but you only have to run this once and it produces the mathematically perfect result. In the past, I got good results with trying to reduce the variance in…

      fxtentacleView on HNParent context

  2. DeepSeek OCR 1 comment

    • That also works purely on text and it's the trick I used in my German speech recognition engine ( https://arxiv.org/abs/2206.12693 ). "I'm studying at Oxford Univ" has basically no loss in meaning even though "University" was truncated to less than half its characters.

      fxtentacleView on HNParent context

  3. Vision Transformers (ViT) Explained 1 comment

    • In this paper https://arxiv.org/abs/2206.12693 I show that choosing the text snippets deliberately so that your choices minimize overall entropy will also minimize result entropy, meaning it'll increase the accuracy of your speech recognition system (for German speech). In my…

      fxtentacleView on HNParent context

  4. Show HN: State-of-the-art German speech recognition in 284 lines of C++ 1 comment

    • This model seems strongly overtrained on CV test set. Usually improvement from LM rescoring is just 10% relative. In the paper https://arxiv.org/pdf/2206.12693.pdf the improvement is from 10.1% WER to 3.64% WER (Table 6). Such a big improvement suggests that LM is biased. Also…

      nshmView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.