Study On The Pronunciation Of Mongolian Names For Mongolian Speech Synthesis System

At present,the Mongolian speech synthesis system has been able to synthesis the text of any length of sentences and paragraphs,and the clarity,intelligibility and naturalness of that have reached a satisfactory level.However,due to the structure,source and pronunciation of the Mongolian name,accurately synthesize the Mongolian name is one of the key issues in the current synthetic system.Based on the establishment of a database of the 28938 Mongolian,this paper uses the methodology of corpus linguistics to study how to correctly machine the Mongolian name in the synthesis system.The full text consists of five parts:introduction,main body(three chapters),conclusions and references.The introduction part mainly summarizes the past research overview of Mongolian names,the reasons for choosing the topic,the meaning of the topic and the research methods.The first chapter expounds the process of establishing the Mongolian database and the problems encountered when restoring the names of the people.The second chapter briefly introduces the general situation of Mongolian speech synthesis research,and discusses in detail the influence of special writing and special reading of Mongolian names on Mongolian speech synthesis system.The third chapter briefly introduces the Mongolian speech synthesis platform as well as classifing and analyzing the synthesis of Mongolian names.In the conclusion part,the research contents of this thesis are summarized and the related contents that are not covered in this paper are pointed out,which will be further studied in the future.