Back to Search

DNN-based speaker-adaptive postfiltering with limited adaptation data for statistical speech synthesis systems

Title	DNN-based speaker-adaptive postfiltering with limited adaptation data for statistical speech synthesis systems
Author	Öztürk, M. G., Ulusoy, O., Demiroğlu, Cenk
Publication Date:	2019
Publication Place	- IEEE
Subject	Speaker adaptation, Speech synthesis, Postfilter, Deep learning
Type	Document
Language	English
Digital	Yes
Manuscript	No
Library:	Özyeğin University
Library Asset ID	978-1-4799-8131-1
Record ID	e71c62e2-a7a5-42e2-9f26-22a7d09970d2
Library Location	Electrical & Electronics Engineering
Date	2019
Notes	TÜBİTAK
Sample Text	Deep neural networks (DNNs) have been successfully deployed for acoustic modelling in statistical parametric speech synthesis (SPSS) systems. Moreover, DNN-based postfilters (PF) have also been shown to outperform conventional postfilters that are widely used in SPSS systems for increasing the quality of synthesized speech. However, existing DNN-based postfilters are trained with speaker-dependent databases. Given that SPSS systems can rapidly adapt to new speakers from generic models, there is a need for DNN-based postfilters that can adapt to new speakers with minimal adaptation data. Here, we compare DNN-, RNN-, and CNN-based postfilters together with adversarial (GAN) training and cluster-based initialization (CI) for rapid adaptation. Results indicate that the feedforward (FF) DNN, together with GAN and CI, significantly outperforms the other recently proposed postfilters.
DOI	10.1109/ICASSP.2019.8683714

View in source Özyeğin University Özyeğin University - Historical works, archives, and periodicals search engine

Özyeğin University

DNN-based speaker-adaptive postfiltering with limited adaptation data for statistical speech synthesis systems

Author Öztürk, M. G., Ulusoy, O., Demiroğlu, Cenk

Publication Date 2019

Publication Place - IEEE

Subject Speaker adaptation, Speech synthesis, Postfilter, Deep learning

Type Document

Language English

Digital Yes

Manuscript No

Library Özyeğin University

Library Asset ID 978-1-4799-8131-1

Record ID e71c62e2-a7a5-42e2-9f26-22a7d09970d2

Library Location Electrical & Electronics Engineering

Date 2019

Notes TÜBİTAK

Sample Text Deep neural networks (DNNs) have been successfully deployed for acoustic modelling in statistical parametric speech synthesis (SPSS) systems. Moreover, DNN-based postfilters (PF) have also been shown to outperform conventional postfilters that are widely used in SPSS systems for increasing the quality of synthesized speech. However, existing DNN-based postfilters are trained with speaker-dependent databases. Given that SPSS systems can rapidly adapt to new speakers from generic models, there is a need for DNN-based postfilters that can adapt to new speakers with minimal adaptation data. Here, we compare DNN-, RNN-, and CNN-based postfilters together with adversarial (GAN) training and cluster-based initialization (CI) for rapid adaptation. Results indicate that the feedforward (FF) DNN, together with GAN and CI, significantly outperforms the other recently proposed postfilters.

DOI 10.1109/ICASSP.2019.8683714