TTS skins: Speaker conversion via ASR

Adam Polyak; Lior Wolf; Yaniv Taigman

doi:10.21437/Interspeech.2020-1416

TTS skins: Speaker conversion via ASR

Adam Polyak^*, Lior Wolf, Yaniv Taigman

^*Corresponding author for this work

School of Computer Science

Research output: Contribution to journal › Conference article › peer-review

Abstract

We present a fully convolutional wav-to-wav network for converting between speakers' voices, without relying on text. Our network is based on an encoder-decoder architecture, where the encoder is pre-trained for the task of Automatic Speech Recognition, and a multi-speaker waveform decoder is trained to reconstruct the original signal in an autoregressive manner. We train the network on narrated audiobooks, and demonstrate multi-voice TTS in those voices, by converting the voice of a TTS robot.

Original language	English
Pages (from-to)	786-790
Number of pages	5
Journal	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume	2020-October
DOIs	https://doi.org/10.21437/Interspeech.2020-1416
State	Published - 2020
Event	21st Annual Conference of the International Speech Communication Association, INTERSPEECH 2020 - Shanghai, China Duration: 25 Oct 2020 → 29 Oct 2020

Keywords

Human-computer interaction
Text to speech
Voice conversion

Access to Document

10.21437/Interspeech.2020-1416

Cite this

@article{5af14f32c77244138d29d0bf778eda7e,

title = "TTS skins: Speaker conversion via ASR",

abstract = "We present a fully convolutional wav-to-wav network for converting between speakers' voices, without relying on text. Our network is based on an encoder-decoder architecture, where the encoder is pre-trained for the task of Automatic Speech Recognition, and a multi-speaker waveform decoder is trained to reconstruct the original signal in an autoregressive manner. We train the network on narrated audiobooks, and demonstrate multi-voice TTS in those voices, by converting the voice of a TTS robot.",

keywords = "Human-computer interaction, Text to speech, Voice conversion",

author = "Adam Polyak and Lior Wolf and Yaniv Taigman",

note = "Publisher Copyright: Copyright {\textcopyright} 2020 ISCA; null ; Conference date: 25-10-2020 Through 29-10-2020",

year = "2020",

doi = "10.21437/Interspeech.2020-1416",

language = "אנגלית",

volume = "2020-October",

pages = "786--790",

journal = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",

issn = "2308-457X",

}

TY - JOUR

T1 - TTS skins

AU - Polyak, Adam

AU - Wolf, Lior

AU - Taigman, Yaniv

PY - 2020

Y1 - 2020

N2 - We present a fully convolutional wav-to-wav network for converting between speakers' voices, without relying on text. Our network is based on an encoder-decoder architecture, where the encoder is pre-trained for the task of Automatic Speech Recognition, and a multi-speaker waveform decoder is trained to reconstruct the original signal in an autoregressive manner. We train the network on narrated audiobooks, and demonstrate multi-voice TTS in those voices, by converting the voice of a TTS robot.

AB - We present a fully convolutional wav-to-wav network for converting between speakers' voices, without relying on text. Our network is based on an encoder-decoder architecture, where the encoder is pre-trained for the task of Automatic Speech Recognition, and a multi-speaker waveform decoder is trained to reconstruct the original signal in an autoregressive manner. We train the network on narrated audiobooks, and demonstrate multi-voice TTS in those voices, by converting the voice of a TTS robot.

KW - Human-computer interaction

KW - Text to speech

KW - Voice conversion

UR - http://www.scopus.com/inward/record.url?scp=85098200006&partnerID=8YFLogxK

U2 - 10.21437/Interspeech.2020-1416

DO - 10.21437/Interspeech.2020-1416

M3 - מאמר מכנס

AN - SCOPUS:85098200006

VL - 2020-October

SP - 786

EP - 790

JO - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

JF - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

SN - 2308-457X

Y2 - 25 October 2020 through 29 October 2020

ER -

TTS skins: Speaker conversion via ASR

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this