arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.15722cs.SDcs.CLcs.LGeess.AS
收藏

基于变换器的带剩余高斯自注意的端到端语音识别

Transformer-based end-to-end speech recognition with residual Gaussian-based self-attention

Chengdong Liang, Menglong Xu, Xiao-Lei Zhang

摘要:

自注意(SA)是一种根据向量序列的两两相似性对向量序列进行编码的方法,由于其较强的上下文建模能力而被广泛应用于语音识别中。然而,当应用于长序列数据时,其精度降低。这是由于其加权平均算子可能导致注意力分布的分散,从而导致忽略相邻信号之间的关系。为了解决这个问题,在本文中,我们引入相对位置意识自我注意(RPSA)。它既保持了自注意的全局-范围依赖建模能力,又提高了局部性建模能力。由于原始RPSA的局部窗口长度是固定的,并且对不同的测试数据敏感,本文提出了一种窗口长度可学习的自适应测试数据的高斯自注意算法(Gaussian-based self-attention,GSA)。我们进一步推广GSA到一个新的残差高斯自注意(resGSA)的性能改善。我们分别将RPSA、GSA和resGSA应用到基于transformer的语音识别中。在AISHELL-1汉语语音识别语料库上的实验结果表明了所提方法的有效性。例如,resGSA-Transformer在测试集上实现了5.86%的字符出错率(CER),这比SA-Transformer的错误率相对低7.8%。虽然所提出的resGSA-Transformer的性能仅略优于RPSA-Transformer,但它不必手动调整窗口长度。

英文摘要:

Self-attention (SA), which encodes vector sequences according to their pairwise similarity, is widely used in speech recognition due to its strong context modeling ability. However, when applied to long sequence data, its accuracy is reduced. This is caused by the fact that its weighted average operator may lead to the dispersion of the attention distribution, which results in the relationship between adjacent signals ignored. To address this issue, in this paper, we introduce relative-position-awareness self-attention (RPSA). It not only maintains the global-range dependency modeling ability of self-attention, but also improves the localness modeling ability. Because the local window length of the original RPSA is fixed and sensitive to different test data, here we propose Gaussian-based self-attention (GSA) whose window length is learnable and adaptive to the test data automatically. We further generalize GSA to a new residual Gaussian self-attention (resGSA) for the performance improvement. We apply RPSA, GSA, and resGSA to Transformer-based speech recognition respectively. Experimental results on the AISHELL-1 Mandarin speech recognition corpus demonstrate the effectiveness of the proposed methods. For example, the resGSA-Transformer achieves a character error rate (CER) of 5.86% on the test set, which is relative 7.8% lower than that of the SA-Transformer. Although the performance of the proposed resGSA-Transformer is only slightly better than that of the RPSA-Transformer, it does not have to tune the window length manually.

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。

↑