LectureRechercheRadarCadre d'investissement
Connexion / Inscription
← Retour à la lecture
ActualitésSujet · BuzzASR模型集Prise rapide1 min de lectureRédaction DisconfirmAI

BuzzASR Boosts Low-Resource Speech Accuracy

New model suite beats Whisper in 77 of 102 languages, cutting error rates by 2.8x.

2026-10-06 12:00 GMT+8

The BuzzASR model suite outperforms Whisper-large-v3 in 77 of 102 languages, reducing the average Character Error Rate (CER) by more than 2.8 times.

Mainstream end-to-end speech recognition models are typically trained on multiple languages, which often leads to poor performance in low-resource languages with limited training data. While monolingual fine-tuning is known to be effective, it had previously been applied only to a small number of languages.

This study scales the monolingual fine-tuning strategy to 102 languages and introduces tokenizer replacement, improving compression rates by an average of 3.3 times. All models, code, and detailed results have been open-sourced.

Sources

  • arxiv.org↗
Sujet

BuzzASR模型集

BuzzASR是一个包含102种语言的单语自动语音识别(ASR)模型集,由Lemn Lab与EleutherAI合作开发。该团队通过微调Whisper-large-v3并替换原生分词器,解决了多语言大模型在小语种上表现不佳的问题。在FLEU…

3 étapes1 piècesMis à jour 2026-10-08 10:15

Comment on en est arrivé là

  1. 2026-09-09
Connexion / Inscription
LectureRechercheRadarCadre d'investissement
论文《BuzzASR: A Swarm of 100+ Monolingual Speech Recognition Models》提交至arXiv (v1)。
arxiv.org ↗
  • 2026-10-04论文修订版 (v2) 发布,确认被EMNLP 2026接收。arxiv.org ↗
  • 2026-10-06新闻指出新模型集在102种语言中77种超越Whisper,平均错误率降2.8倍。arxiv.org ↗
  • Ce qui reste non résolu

    • BuzzASR模型在非FLEURS/Common Voice领域的泛化能力如何?
    • 单语模型部署带来的维护成本与多语言统一模型相比是否更具优势?
    Lire le sujet complet →
    Continuer la recherche →