arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.03247eess.AScs.CVcs.SD
收藏

基于ASR的Transformer说话人自适应方法研究

Investigation of Speaker-adaptation methods in Transformer based ASR

Indian Institute of Technology Madras(印度理工学院马德拉斯分校)

Vishwas M. Shetty, Metilda Sagaya Mary N J, S. Umesh

摘要:

在自动语音识别中,端到端模型正在迅速取代传统的混合模型。Transformer是一种基于自注意的序列到序列模型,广泛应用于机器翻译任务中,当用于自动语音识别时,已经给出了有希望的结果。本文探讨了在编码器输入端结合说话人信息的不同方法,同时训练基于变换器的模型,以提高其语音识别性能。我们目前的扬声器信息的形式扬声器嵌入每个扬声器。我们实验使用两种类型的扬声器嵌入:x向量和新的s向量在我们以前的工作中提出的。我们报告了两个数据集的结果a)NPTEL讲座数据库和b)Librispeech 500小时分裂。NPTEL是一个开源的电子学习门户网站,提供来自印度顶尖大学的讲座。我们通过将说话人嵌入到模型中的方法,在基线上提高了单词错误率。

英文摘要:

End-to-end models are fast replacing the conventional hybrid models in automatic speech recognition. Transformer, a sequence-to-sequence model, based on self-attention popularly used in machine translation tasks, has given promising results when used for automatic speech recognition. This paper explores different ways of incorporating speaker information at the encoder input while training a transformer-based model to improve its speech recognition performance. We present speaker information in the form of speaker embeddings for each of the speakers. We experiment using two types of speaker embeddings: x-vectors and novel s-vectors proposed in our previous work. We report results on two datasets a) NPTEL lecture database and b) Librispeech 500-hour split. NPTEL is an open-source e-learning portal providing lectures from top Indian universities. We obtain improvements in the word error rate over the baseline through our approach of integrating speaker embeddings into the model.

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。

↑