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 1909.09116cs.CLcs.LGeess.AS
收藏

端到端语音识别的自训练

Self-Training for End-to-End Speech Recognition

Facebook AI Research(脸书人工智能研究院)

Jacob Kahn, Ann Lee, Awni Hannun

摘要:

我们在端到端语音识别的背景下重新审视自我训练。我们证明,使用伪标签进行训练可以大大提高基线模型的准确性。我们的方法的关键是一个强大的基线声学和语言模型用于生成伪标签,过滤机制定制的序列到序列模型的常见错误,和一种新的集成方法,以增加伪标签的多样性。LibriSpeech语料库上的实验表明,与在嘈杂的语音设置中对100小时的标记数据进行训练的基线相比,使用四个模型和标签过滤的集成,自训练在WER中产生33.9%的相对改善。在干净的语音设置中,自训练恢复了基线和oracle模型之间的59.3%的差距,这比以前的方法至少高出93.8%。

英文摘要:

We revisit self-training in the context of end-to-end speech recognition. We demonstrate that training with pseudo-labels can substantially improve the accuracy of a baseline model. Key to our approach are a strong baseline acoustic and language model used to generate the pseudo-labels, filtering mechanisms tailored to common errors from sequence-to-sequence models, and a novel ensemble approach to increase pseudo-label diversity. Experiments on the LibriSpeech corpus show that with an ensemble of four models and label filtering, self-training yields a 33.9% relative improvement in WER compared with a baseline trained on 100 hours of labelled data in the noisy speech setting. In the clean speech setting, self-training recovers 59.3% of the gap between the baseline and an oracle model, which is at least 93.8% relatively higher than what previous approaches can achieve.

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。

↑