论文部分内容阅读
通过语音输入方式准确识别蒙古语主要方言 ,并为其建立共享识别模型 ,对实现方言间的信息自动交换及语音识别技术的实用化具有实际意义。针对蒙古语多方言的语音特点 ,可以采取独立的或者共享识别模型进行识别。本文介绍的抽出初步模型和建立共享模型的方法 ,简便易行 ,但识别精度有待提高。文中探索了通过借用数据量和质较完备的语言的语音数据提高识别率的可能性。
It is of practical significance to recognize the main Mongolian dialects through voice input and to establish a shared recognition model for them. It is of practical significance to realize the automatic exchange of information between dialects and the practical application of speech recognition technology. For Mongolian multi-dialect voice features, you can take an independent or shared identification model to identify. This article describes the extraction of the initial model and the establishment of a shared model method, easy, but the recognition accuracy needs to be improved. The paper explores the possibility of improving the recognition rate by borrowing the data of the voice and the quality of speech.