Unsupervised Polyglot Text-to-speech

Eliya Nachmani; Lior Wolf

doi:10.1109/ICASSP.2019.8683519

Unsupervised Polyglot Text-to-speech

Eliya Nachmani, Lior Wolf

Tel Aviv University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

We present a TTS neural network that is able to produce speech in multiple languages. The proposed network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages. Training is done without using matching or parallel data, i.e., without samples of the same speaker in multiple languages, making the method much more applicable. The conversion is based on learning a polyglot network that has multiple per-language sub-networks and adding loss terms that preserve the speaker's identity in multiple languages. We evaluate the proposed polyglot neural network for three languages with a total of more than 400 speakers and demonstrate convincing conversion capabilities.

Original language	American English
Title of host publication	2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Pages	7055-7059
Number of pages	5
ISBN (Electronic)	9781479981311
DOIs	https://doi.org/10.1109/ICASSP.2019.8683519
State	Published - 1 May 2019
Event	44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, United Kingdom Duration: 12 May 2019 → 17 May 2019

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2019-May

Conference

Conference	44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
Country/Territory	United Kingdom
City	Brighton
Period	12/05/19 → 17/05/19

Keywords

TTS
multilingual
unsupervised learning

All Science Journal Classification (ASJC) codes

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2019.8683519

Cite this

@inproceedings{6d0d84316fd343ada3a3bbe06d796577,

title = "Unsupervised Polyglot Text-to-speech",

abstract = "We present a TTS neural network that is able to produce speech in multiple languages. The proposed network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages. Training is done without using matching or parallel data, i.e., without samples of the same speaker in multiple languages, making the method much more applicable. The conversion is based on learning a polyglot network that has multiple per-language sub-networks and adding loss terms that preserve the speaker's identity in multiple languages. We evaluate the proposed polyglot neural network for three languages with a total of more than 400 speakers and demonstrate convincing conversion capabilities.",

keywords = "TTS, multilingual, unsupervised learning",

author = "Eliya Nachmani and Lior Wolf",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 ; Conference date: 12-05-2019 Through 17-05-2019",

year = "2019",

month = may,

day = "1",

doi = "10.1109/ICASSP.2019.8683519",

language = "American English",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "7055--7059",

booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

}

Nachmani, E & Wolf, L 2019, Unsupervised Polyglot Text-to-speech. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8683519, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, pp. 7055-7059, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, Brighton, United Kingdom, 12/05/19. https://doi.org/10.1109/ICASSP.2019.8683519

Unsupervised Polyglot Text-to-speech. / Nachmani, Eliya ; Wolf, Lior.
2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. 2019. p. 7055-7059 8683519 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Unsupervised Polyglot Text-to-speech

AU - Nachmani, Eliya

AU - Wolf, Lior

PY - 2019/5/1

Y1 - 2019/5/1

N2 - We present a TTS neural network that is able to produce speech in multiple languages. The proposed network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages. Training is done without using matching or parallel data, i.e., without samples of the same speaker in multiple languages, making the method much more applicable. The conversion is based on learning a polyglot network that has multiple per-language sub-networks and adding loss terms that preserve the speaker's identity in multiple languages. We evaluate the proposed polyglot neural network for three languages with a total of more than 400 speakers and demonstrate convincing conversion capabilities.

AB - We present a TTS neural network that is able to produce speech in multiple languages. The proposed network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages. Training is done without using matching or parallel data, i.e., without samples of the same speaker in multiple languages, making the method much more applicable. The conversion is based on learning a polyglot network that has multiple per-language sub-networks and adding loss terms that preserve the speaker's identity in multiple languages. We evaluate the proposed polyglot neural network for three languages with a total of more than 400 speakers and demonstrate convincing conversion capabilities.

KW - TTS

KW - multilingual

KW - unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85068988043&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8683519

DO - 10.1109/ICASSP.2019.8683519

M3 - Conference contribution

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 7055

EP - 7059

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

T2 - 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019

Y2 - 12 May 2019 through 17 May 2019

ER -

Unsupervised Polyglot Text-to-speech

Abstract

Publication series

Conference

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this