Initial investigation of an encoder-decoder end-to-end TTS framework using marginalization of monotonic hard latent alignments

Yusuke Yasuda, Xin Wang, Junichi Yamagishi

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

Abstract

End-to-end text-to-speech (TTS) synthesis is a method that directly converts input text to output acoustic features using a single network. A recent advance of end-to-end TTS is due to a key technique called attention mechanisms, and all successful methods proposed so far have been based on soft attention mechanisms. However, although network structures are becoming increasingly complex, end-to-end TTS systems with soft attention mechanisms may still fail to learn and to predict accurate alignment between the input and output. This may be because the soft attention mechanisms are too flexible. Therefore, we propose an approach that has more explicit but natural constraints suitable for speech signals to make alignment learning and prediction of end-to-end TTS systems more robust. The proposed system, with the constrained alignment scheme borrowed from segment-to-segment neural transduction (SSNT), directly calculates the joint probability of acoustic features and alignment given an input text. The alignment is designed to be hard and monotonically increase by considering the speech nature, and it is treated as a latent variable and marginalized during training. During prediction, both the alignment and acoustic features can be generated from the probabilistic distributions. The advantages of our approach are that we can simplify many modules for the soft attention and that we can train the end-to-end TTS model using a single likelihood function. As far as we know, our approach is the first end-to-end TTS without a soft attention mechanism.
Original languageEnglish
Title of host publicationProceedings of the 10th ISCA Speech Synthesis Workshop
PublisherInternational Speech Communication Association
Pages1-6
Number of pages6
DOIs
Publication statusPublished - 22 Sep 2019
EventThe 10th ISCA Speech Synthesis Workshop - Austrian museum of folk life and folk art in Vienna, Vienna, Austria
Duration: 20 Sep 201922 Sep 2019
Conference number: 10
http://ssw10.oeaw.ac.at/index.html

Publication series

NameProc. 10th ISCA Speech Synthesis Workshop
PublisherISCA
ISSN (Electronic)2312-2846

Conference

ConferenceThe 10th ISCA Speech Synthesis Workshop
Abbreviated titleSSW 2019
CountryAustria
CityVienna
Period20/09/1922/09/19
Internet address

Keywords

  • text-to-speech synthesis
  • end-to-end
  • neural network

Fingerprint Dive into the research topics of 'Initial investigation of an encoder-decoder end-to-end TTS framework using marginalization of monotonic hard latent alignments'. Together they form a unique fingerprint.

Cite this