|
|||||
|
|
| 基于端到端的蒙古语异形同音词声学建模方法 | |
| 引用本文： | 陈艳,李图雅,马志强,谢秀兰,王洪彬.基于端到端的蒙古语异形同音词声学建模方法[J].中文信息学报,2022,36(3):27-35. |
| 作者姓名： | 陈艳 李图雅 马志强 谢秀兰 王洪彬 |
| 作者单位： | 1.内蒙古工业大学 数据科学与应用学院,内蒙古 呼和浩特 010080; 2.内蒙古工业大学 内蒙古自治区基于大数据的软件服务工程技术研究中心,内蒙古 呼和浩特 010080 |
| 基金项目： | 国家自然科学基金 (61762070,61862048);内蒙古自治区自然科学基金 (2019MS06004);内蒙古自治区科技重大专项(2019ZD015);内蒙古自治区关键技术攻关计划项目(2019GG273) |
| 摘 要： | 蒙古语声学模型的训练过程是模型学习发音数据与标注数据之间关系的一个过程.针对以音素为建模粒子的蒙古语声学模型建模,由于蒙古语词的发音与语义存在一对多映射现象,会造成解码出的蒙古语文本错误,进而导致蒙古语语音识别系统识别率降低的问题.对此,该文以端到端模型为基础,以蒙古语音素、字母为蒙古语声学模型建模粒子,设计了基于BL... |
| 关 键 词： | 异形同音词 建模粒子 端到端 蒙古语声学模型 语音识别 |
An End-to-End Acoustic Modeling Approach to Mongolian Heteromorphic Homophones |
|
| CHEN Yan,LI Tuya,MA Zhiqiang,XIE Xiulan,WANG Hongbin.An End-to-End Acoustic Modeling Approach to Mongolian Heteromorphic Homophones[J].Journal of Chinese Information Processing,2022,36(3):27-35. | |
| Authors: | CHEN Yan LI Tuya MA Zhiqiang XIE Xiulan WANG Hongbin |
| Affiliation: | 1.College of Data Science and Application, Inner Mongolia University of Technology, Hohhot, Inner Mongolia 010080, China; 2.Inner Mongolia Autonomous Region Engineering and Technology Research Centre of Big Data Based Software Service, Inner Mongolia University of Technology, Hohhot, Inner Mongolia 010080, China |
| Abstract: | The training process of Mongolian acoustic model is a process where the model learns the relationship between pronunciation data and annotation data. Aiming at the modeling of Mongolianacoustic model based on phonemes, deal with the one-to-many mapping phenomenon between pronunciation and semantics, which will the decoding of Mongolian text will be wrong and will lead to the problem of low recognition rate of Mongolian speech recognition system. In this regard, this paper designs an End-to-End Mongolian acoustic model with both phonemes and letters used. Specifically, a Mongolian acoustic model based on BLSTM-CTC is described, and a momentum training algorithm is applied. The experimental results show that the proposed method can effectively reduce the word error rate of heteromorphous homophones in Mongolian speech recognition system. |
| Keywords: | heteromorphic homophone modeling unit End-to-End Mongolian acoustic model speech recognition |
| 点击此处可从《中文信息学报》浏览原始摘要信息 | |
| 点击此处可从《中文信息学报》下载全文 | |