期刊
INTERSPEECH 2022
卷 -, 期 -, 页码 3198-3202出版社
ISCA-INT SPEECH COMMUNICATION ASSOC
DOI: 10.21437/Interspeech.2022-10109
关键词
homophone disambiguation; prosodic features; Random Forest; CNN; conversational speech; Austrian German
类别
资金
- Austrian Science Fund (FWF) [V-638-N33]
The high degree of segmental reduction in conversational speech leads to a large number of words becoming homophones, which poses a challenge for automatic speech recognition. This study proposes two approaches, one based on prosodic and spectral features and the other based on a convolutional neural network, to disambiguate homophones. The results show potential for both approaches, especially when combined with a stochastic language model as part of an ASR system.
Given the high degree of segmental reduction in conversational speech, a large number of words become homophoneous that in read speech are not. For instance, the tokens considered in this study ah, ach, auch, eine and er may all be reduced to [a] in conversational Austrian German. Homophones pose a serious problem for automatic speech recognition (ASR), where homophone disambiguation is typically solved using lexical context. In contrast, we propose two approaches to disambiguate homophones on the basis of prosodic and spectral features. First, we build a Random Forest classifier with a large set of acoustic features, which reaches good performance given the small data size, and allows us to gain insight into how these homophones are distinct with respect to phonetic detail. Since for the extraction of the features annotations are required, this approach would not be practical for the integration into an ASR system. We thus explored a second, convolutional neural network (CNN) based approach. The performance of this approach is on par with the one based on Random Forest, and the results indicate a high potential of this approach to facilitate homophone disambiguation when combined with a stochastic language model as part of an ASR system.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据