Please use this identifier to cite or link to this item:
https://open.uns.ac.rs/handle/123456789/2043
Title: | Style-code method for multi-style parametric text-to-speech syntesis | Authors: | Suzić, Siniša Delić, Tijana Ostrogonac S. Đuričić , Jasna Pekar, Darko |
Issue Date: | 1-Jan-2018 | Journal: | SPIIRAS Proceedings | Abstract: | © 2018 St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences. All rights reserved. Modern text-to-speech systems generally achieve good intelligibility. The one of the main drawbacks of these systems is the lack of expressiveness in comparison to natural human speech. It is very unpleasant when automated system conveys positive and negative message in completely the same way. The introduction of parametric methods in speech synthesis gave possibility to easily change speaker characteristics and speaking styles. In this paper a simple method for incorporating styles into synthesized speech by using style codes is presented. The proposed method requires just a couple of minutes of target style and moderate amount of neutral speech. It is successfully applied to both hidden Markov models and deep neural networks-based synthesis, giving style code as additional input to the model. Listening tests confirmed that better style expressiveness is achieved by deep neural networks synthesis compared to hidden Markov model synthesis. It is also proved that quality of speech synthesized by deep neural networks in a certain style is comparable with the speech synthesized in neutral style, although the neutral-speech-database is about 10 times bigger. DNN based TTS with style codes are further investigated by comparing the quality of speech produced by single-style modeling and multi-style modeling systems. Objective and subjective measures confirmed that there is no significant difference between these two approaches. | URI: | https://open.uns.ac.rs/handle/123456789/2043 | ISSN: | 20789181 | DOI: | 10.15622/sp.60.8 |
Appears in Collections: | FTN Publikacije/Publications |
Show full item record
SCOPUSTM
Citations
2
checked on May 20, 2023
Page view(s)
25
Last Week
10
10
Last month
0
0
checked on May 10, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.