Integrating prosodic information into recurrent neural network language model for speech recognition | |
Fu, Tong ; Han, Yang ; Li, Xiangang ; Liu, Yi ; Wu, Xihong | |
2015 | |
英文摘要 | Prosody is a kind of cues that are critical to human speech perception and comprehension, so it is plausible to integrate prosodic information into machine speech recognition. However, as a result of the supra-segmental nature, it is hard to integrate prosodic information with conventional acoustic features. Recently, RNNLMs have shown to be the state-of-the-art language model in many tasks. We thus attempt to integrate prosodic information into RNNLMs for improving speech recognition performance based on rescoring strategy. Firstly, three word-level prosodic features are extracted from speech and then passed to RNNLMs separately. Therefore RNNLMs predict the next word based on prosodic features and word history. Experiments conducted on LibriSpeech Corpus show that the word error rate decreases from 8.07% to 7.96%. Secondly, prosodic information is combined on feature-level and model-level for further improvements and word error rate decreases 4.71% relatively. ? 2015 Asia-Pacific Signal and Information Processing Association.; EI; 1194-1197 |
语种 | 英语 |
出处 | 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015 |
DOI标识 | 10.1109/APSIPA.2015.7415462 |
内容类型 | 其他 |
源URL | [http://ir.pku.edu.cn/handle/20.500.11897/449526] ![]() |
专题 | 信息科学技术学院 |
推荐引用方式 GB/T 7714 | Fu, Tong,Han, Yang,Li, Xiangang,et al. Integrating prosodic information into recurrent neural network language model for speech recognition. 2015-01-01. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论