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 2206.11596eess.AScs.AI
收藏

基于两遍解码和交叉自适应的端到端Conformer与混合TDNN语音识别系统组合

Two-pass Decoding and Cross-adaptation Based System Combination of End-to-end Conformer and Hybrid TDNN ASR Systems

The Chinese University of Hong Kong(香港中文大学);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所)

Mingyu Cui, Jiajun Deng, Shoukang Hu, Xurong Xie, Tianzi Wang, Shujie Hu, Mengzhe Geng, Boyang Xue, Xunying Liu, Helen Meng

AI总结:

本文提出多遍重打分与交叉自适应的系统组合方法,融合混合TDNN与Conformer E2E ASR系统,在Switchboard上实现显著WER降低,最佳组合系统相对Conformer绝对降低2.5%-3.9%。

摘要:

混合自动语音识别(ASR)系统与端到端(E2E)ASR系统在基本建模方式上的差异产生了它们之间较大的多样性和互补性。本文研究了基于多遍重打分和交叉自适应的系统组合方法,用于混合TDNN和Conformer E2E ASR系统。在多遍重打分中,采用最先进的混合LF-MMI训练的CNN-TDNN系统,该系统具有速度扰动、SpecAugment和贝叶斯学习隐藏单元贡献(LHUC)说话人自适应,用于生成初始的N-best输出,然后由说话人自适应的Conformer系统使用2路交叉系统分数插值进行重打分。在交叉自适应中,混合CNN-TDNN系统自适应于Conformer系统的1-best输出,反之亦然。在300小时Switchboard语料库上的实验表明,使用这两种系统组合方法中的任何一种得到的组合系统都优于单个系统。使用多遍重打分获得的最佳组合系统在NIST Hub5'00、Rt03和Rt02评估数据上,与单独的Conformer系统相比,实现了统计显著的词错误率(WER)绝对降低2.5%至3.9%(相对降低22.5%至28.9%)。

英文摘要:

Fundamental modelling differences between hybrid and end-to-end (E2E) automatic speech recognition (ASR) systems create large diversity and complementarity among them. This paper investigates multi-pass rescoring and cross adaptation based system combination approaches for hybrid TDNN and Conformer E2E ASR systems. In multi-pass rescoring, state-of-the-art hybrid LF-MMI trained CNN-TDNN system featuring speed perturbation, SpecAugment and Bayesian learning hidden unit contributions (LHUC) speaker adaptation was used to produce initial N-best outputs before being rescored by the speaker adapted Conformer system using a 2-way cross system score interpolation. In cross adaptation, the hybrid CNN-TDNN system was adapted to the 1-best output of the Conformer system or vice versa. Experiments on the 300-hour Switchboard corpus suggest that the combined systems derived using either of the two system combination approaches outperformed the individual systems. The best combined system obtained using multi-pass rescoring produced statistically significant word error rate (WER) reductions of 2.5% to 3.9% absolute (22.5% to 28.9% relative) over the stand alone Conformer system on the NIST Hub5'00, Rt03 and Rt02 evaluation data.

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。

↑