Paper Info
Title
Recognition for synthesis: Automatic parameter selection for resynthesis of emotional speech from neutral speech
Abstract
One of the biggest challenges in emotional speech resynthesis is the selection of modification parameters that will make humans perceive a targeted emotion. The best selection method is by using human raters. However, for large evaluation sets this process can be very costly. In this paper, we describe a recognition for synthesis (RFS) system to automatically select a set of possible parameter values that can be used to resynthesize emotional speech. The system, developed with supervised training, consists of synthesis (TD-PSOLA), recognition (neural network) and parameter selection modules. The experimental results show evidence that the parameter sets selected by the RFS system can be successfully used to resynthesize the input neutral speech as angry speech, demonstrating that the RFS system can assist in the human evaluation of emotional speech.
Year
DOI
Venue
2008
10.1109/ICASSP.2008.4518688
Las Vegas, NV
Keywords
Field
DocType
emotion recognition,neural nets,speech recognition,speech synthesis,TD-PSOLA,automatic parameter selection,emotional speech synthesis,neural network,neutral speech,speech recognition,supervised training,automatic evaluation,emotion resynthesis,neural network,recognition for synthesis
Speech synthesis,Computer science,Emotion recognition,Speech recognition,Natural language processing,Artificial intelligence,Supervised training,Artificial neural network
Conference
ISSN
ISBN
Citations 
1520-6149 E-ISBN : 978-1-4244-1484-0
978-1-4244-1484-0
6
PageRank 
References 
Authors
0.60
4
3
Name
Order
Citations
PageRank
Murtaza Bulut184839.45
Sungbok Lee2139484.13
Narayanan Shrikanth35558439.23