TEVR: Improving Speech Recognition by Token Entropy Variance Reduction
Hajo Nils Krabbenhöft, Erhardt Barth
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/DeutscheKI/tevr-asr-toolOfficialnone★ 412
- huggingface.co/fxtentacle/wav2vec2-xls-r-1b-tevrOfficialnone★ 0
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.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| Common Voice German | wav2vec 2.0 XLS-R 1B + TEVR (5-gram) | Test WER | 3.64 | — | Unverified |
| Common Voice German | wav2vec 2.0 XLS-R 1B + TEVR (4-gram) | Test WER | 3.7 | — | Unverified |
| Common Voice German | wav2vec 2.0 XLS-R 1B (5-gram) | Test WER | 4.38 | — | Unverified |
| Common Voice German | wav2vec 2.0 XLS-R 1B + TEVR (no LM) | Test WER | 10.1 | — | Unverified |
| Common Voice German | wav2vec 2.0 XLS-R (no LM) | Test WER | 12.06 | — | Unverified |