arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.09428eess.AScs.LGcs.SD
收藏

基于贝叶斯非自回归模型的流媒体端到端ASR

Streaming End-to-End ASR based on Blockwise Non-Autoregressive Models

Johns Hopkins University(约翰斯·霍普金斯大学);Yahoo Japan Corporation(雅虎日本公司);Carnegie Mellon University(卡内基梅隆大学)

Tianzi Wang, Yuya Fujita, Xuankai Chang, Shinji Watanabe

摘要:

非自回归(NAR)建模在语音处理中得到了越来越多的关注。与最近的国家的最先进的注意力为基础的自动语音识别(ASR)的结构,NAR可以实现有前途的实时因子(RTF)的改善,只有很小的退化的准确性相比,自回归(AR)模型。然而,识别推理需要等待完整的语音话语的完成,这限制了它们在低延迟场景中的应用。为了解决这个问题,我们提出了一种新的端到端的流NAR语音识别系统相结合的blockwise-attention和连接主义的时间分类与掩码预测(Mask-CTC)NAR。在推理过程中,输入音频被分成小块,然后以分块流式方式进行处理。为了解决每个块输出边缘的插入和删除错误,我们采用了一种重叠解码策略,并采用了一种动态映射技巧,可以产生更连贯的句子。实验结果表明,与vanilla Mask-CTC相比,该方法在低延迟条件下提高了在线ASR识别率。此外,与基于AR注意力的模型相比,它可以实现更快的推理速度。我们所有的代码都将在https://github.com/espnet/espnet上公开。

英文摘要:

Non-autoregressive (NAR) modeling has gained more and more attention in speech processing. With recent state-of-the-art attention-based automatic speech recognition (ASR) structure, NAR can realize promising real-time factor (RTF) improvement with only small degradation of accuracy compared to the autoregressive (AR) models. However, the recognition inference needs to wait for the completion of a full speech utterance, which limits their applications on low latency scenarios. To address this issue, we propose a novel end-to-end streaming NAR speech recognition system by combining blockwise-attention and connectionist temporal classification with mask-predict (Mask-CTC) NAR. During inference, the input audio is separated into small blocks and then processed in a blockwise streaming way. To address the insertion and deletion error at the edge of the output of each block, we apply an overlapping decoding strategy with a dynamic mapping trick that can produce more coherent sentences. Experimental results show that the proposed method improves online ASR recognition in low latency conditions compared to vanilla Mask-CTC. Moreover, it can achieve a much faster inference speed compared to the AR attention-based models. All of our codes will be publicly available at https://github.com/espnet/espnet.

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。

↑