INTERSPEECH 2015
16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

Personalized Synthetic Voices for Speaking Impaired: Website and App

D. Erro (1), Inma Hernaez (1), Agustin Alonso (1), D. García-Lorenzo (1), Eva Navas (1), J. Ye (1), H. Arzelus (2), Igor Jauk (3), N. Q. Hy (4), C. Magariños (5), R. Pérez-Ramón (1), M. Sulír (6), Xiaohai Tian (4), X. Wang (7)

(1) Universidad del País Vasco, Spain
(2) VicomTech-IK4, Spain
(3) Universitat Politècnica de Catalunya, Spain
(4) NTU, Singapore
(5) Universidade de Vigo, Spain
(6) Technical University of Košice, Slovak Republic
(7) USTC, China

This paper describes the current state of the work that is being carried out in the framework of the ZureTTS project to give a personalized voice to people who cannot speak in their own. Despite the availability of tools and algorithms to synthesize speech and adapt it to new speakers, this process is affordable only for experts. To overcome this problem, we recently developed a web interface that assists users in doing so. At this point only healthy users can fully personalize the synthetic voice via adaptation, while impaired users can just manually tune a few dimensions of it. As a complement, we have launched an Android application that connects to the ZureTTS server and makes use of its functionalities in an intuitive way. Although many parts of the system need to be improved, its current version is publicly accessible and ready to be used.

Full Paper

Bibliographic reference.  Erro, D. / Hernaez, Inma / Alonso, Agustin / García-Lorenzo, D. / Navas, Eva / Ye, J. / Arzelus, H. / Jauk, Igor / Hy, N. Q. / Magariños, C. / Pérez-Ramón, R. / Sulír, M. / Tian, Xiaohai / Wang, X. (2015): "Personalized synthetic voices for speaking impaired: website and app", In INTERSPEECH-2015, 1251-1254.