This paper describes the Speect text-to-speech system entry submitted to the Blizzard Challenge 2016. The focus of this entry was to build a data driven text-to-speech system that generates an expressive voice suitable for children’s audio books. The techniques applied for the task of the challenge and the implementation details for the alignment of the audio books and text-to-speech modules are described. The results of the evaluations are given and discussed.
Reference:
Louw, J.A., Moodley, A. & Govender, A. 2016. The Speect text-to-speech entry for the Blizzard Challenge 2016. In: Blizzard Challenge 2016, Cupertino, United States of America, 16 September 2016
Louw, J. A., Moodley, A., & Govender, A. (2016). The Speect text-to-speech entry for the Blizzard Challenge 2016. Blizzard Challenge. http://hdl.handle.net/10204/11323
Louw, Johannes A, Avashlin Moodley, and Avashna Govender. "The Speect text-to-speech entry for the Blizzard Challenge 2016." (2016): http://hdl.handle.net/10204/11323
Louw JA, Moodley A, Govender A, The Speect text-to-speech entry for the Blizzard Challenge 2016; Blizzard Challenge; 2016. http://hdl.handle.net/10204/11323 .
