Edinburgh Research Explorer

Training Multi-Speaker Neural Text-to-Speech Systems using Speaker-Imbalanced Speech Corpora

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

Related Edinburgh Organisations

Original languageEnglish
Title of host publicationProceedings Interspeech 2019
Number of pages5
Publication statusAccepted/In press - 17 Jun 2019
EventInterspeech 2019 - Graz, Austria
Duration: 15 Sep 201919 Sep 2019
https://www.interspeech2019.org/

Conference

ConferenceInterspeech 2019
CountryAustria
CityGraz
Period15/09/1919/09/19
Internet address

Abstract

When the available data of a target speaker is insufficient to train a high quality speaker-dependent neural text-to-speech (TTS) system, we can combine data from multiple speakers and train a multi-speaker TTS model instead. Many studies have shown that neural multi-speaker TTS model trained with a small amount data from multiple speakers combined can generate synthetic speech with better quality and stability than a speaker-dependent one. However when the amount of data from each speaker is highly unbalanced, the best approach to make use of the excessive data remains unknown. Our experiments showed that simply combining all available data from every speaker to train a multi-speaker model produces better than or at least similar performance to its speaker-dependent counterpart. Moreover by using an ensemble multi-speaker model, in which each subsystem is trained on a subset of available data, we can further improve the quality of the synthetic speech especially for underrepresented speakers whose training data is limited.

    Research areas

  • speech synthesis, multi-speaker modeling, imbalanced corpus, ensemble learning

Event

Interspeech 2019

15/09/1919/09/19

Graz, Austria

Event: Conference

ID: 99989925