Improving Trajectory Modelling for DNN-Based Speech Synthesis by Using Stacked Bottleneck Features and Minimum Generation Error Training (2016)
Attributed to:
Natural Speech Technology
funded by
EPSRC
Abstract
No abstract provided
Bibliographic Information
Digital Object Identifier: http://dx.doi.org/10.1109/taslp.2016.2551865
Publication URI: http://dx.doi.org/10.1109/taslp.2016.2551865
Type: Journal Article/Review
Parent Publication: IEEE/ACM Transactions on Audio, Speech, and Language Processing
Issue: 7