Abstract
In recent years, with the development of deep learning, speech synthesis technology has developed rapidly. Limited to the small size of corpus data, the field of Tibetan speech synthesis has several problems, such as slow development and insufficient in-depth research. Therefore, the construction of Tibetan speech synthetic dataset is of great significance. In this paper, to meet the needs of speech synthesis system based on neural network, according to the methods of phoneme balance, Tibetan character frequency analysis, Tibetan character word-formation analysis, we design a speech corpus which abides by the characteristics of Tibetan Amdo dialect. Using professional instruments and equipment, we record two professional hosts Tibetan Amdo audio, and make a 60 hours of Tibetan Amdo dialect speech dataset. In the mainstream end-to-end speech synthesis framework, it can prove that the synthetic speech is natural and understandable, which shows the effectiveness and usefulness of our dataset.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
