arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.06204cs.CL
收藏

基于语音的低资源上下文主题识别

Low-Resource Contextual Topic Identification on Speech

The Johns Hopkins University(约翰斯·霍普金斯大学)

Chunxi Liu, Matthew Wiesner, Shinji Watanabe, Craig Harman, Jan Trmal, Najim Dehak, Sanjeev Khudanpur

摘要:

在现实世界的非结构化音频的主题识别(主题ID)中,可变主题转换的音频实例首先被分解成连续的片段,并且每个片段被独立地分类。我们首先提出了一个通用的方法,在低资源的语言口语段的主题ID,使用级联的通用声学建模,翻译词典到英语,英语主题分类。接下来,而不是独立地分类每个段,我们证明,探索跨顺序段的上下文依赖关系可以提供很大的改进。特别是,我们提出了一个基于注意力的上下文模型,它能够利用上下文中有选择的方式。我们测试我们的上下文和非上下文模型四个LORELEI语言,但我们的注意力为基础的上下文模型显着优于上下文无关的模型。

英文摘要:

In topic identification (topic ID) on real-world unstructured audio, an audio instance of variable topic shifts is first broken into sequential segments, and each segment is independently classified. We first present a general purpose method for topic ID on spoken segments in low-resource languages, using a cascade of universal acoustic modeling, translation lexicons to English, and English-language topic classification. Next, instead of classifying each segment independently, we demonstrate that exploring the contextual dependencies across sequential segments can provide large improvements. In particular, we propose an attention-based contextual model which is able to leverage the contexts in a selective manner. We test both our contextual and non-contextual models on four LORELEI languages, and on all but one our attention-based contextual model significantly outperforms the context-independent models.

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。

↑