DNN-based stochastic postfilter for HMM-based speech synthesis

Ling-Hui Chen, Tuomo Raitio, Cassia Valentini-Botinhao, Junichi Yamagishi, Zhen-Hua Ling

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In this paper we propose a deep neural network to model the conditional probability of the spectral differences between natural and synthetic speech. This allows us to reconstruct the spectral fine structures in speech generated by HMMs. We compared the new stochastic data-driven postfilter with global variance based parameter generation and modulation spectrum enhancement. Our results confirm that the proposed method significantly improves the segmental quality of synthetic speech compared to the conventional methods.
Original languageEnglish
Title of host publicationInterspeech 2014
Place of PublicationSingapore
PublisherInternational Speech Communication Association
Number of pages5
Publication statusPublished - Sep 2014


Dive into the research topics of 'DNN-based stochastic postfilter for HMM-based speech synthesis'. Together they form a unique fingerprint.

Cite this