Transforming F0 Contours

Ben Gillett; Simon King

Transforming F0 Contours

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

Abstract

Voice transformation is the process of transforming the characteristics of speech uttered by a source speaker, such that a listener would believe the speech was uttered by a target speaker. Training F0 contour generation models for speech synthesis requires a large corpus of speech. If it were possible to adapt the F0 contour of one speaker to sound like that of another speaker, using a small, easily obtainable parameter set, this would be extremely valuable. We present a new method for the transformation of F0 contours from one speaker to another based on a small linguistically motivated parameter set. The system performs a piecewise linear mapping using these parameters. A perceptual experiment clearly demonstrates that the presented system is at least as good as an existing technique for all speaker pairs, and that in many cases it is much better and almost as good as using the target F0 contour

Original language	English
Title of host publication	Eurospeech 2003 - Interspeech 2003
Subtitle of host publication	8th European Conference on Speech Communication and Technology
Publisher	International Speech Communication Association
Pages	101-104
Number of pages	4
ISBN (Print)	ISSN: 1990-9772
Publication status	Published - 1 Sept 2003

Access to Document

Poster: ben_F0Accepted author manuscript, 146 KB

Concatenative Speech synthesis using articulatory information.
King, S. (Principal Investigator)
EPSRC
1/07/02 → 30/06/05
Project: Research

Cite this

@inproceedings{f23cc8420fb14e9f92573b4095595daf,

title = "Transforming F0 Contours",

abstract = "Voice transformation is the process of transforming the characteristics of speech uttered by a source speaker, such that a listener would believe the speech was uttered by a target speaker. Training F0 contour generation models for speech synthesis requires a large corpus of speech. If it were possible to adapt the F0 contour of one speaker to sound like that of another speaker, using a small, easily obtainable parameter set, this would be extremely valuable. We present a new method for the transformation of F0 contours from one speaker to another based on a small linguistically motivated parameter set. The system performs a piecewise linear mapping using these parameters. A perceptual experiment clearly demonstrates that the presented system is at least as good as an existing technique for all speaker pairs, and that in many cases it is much better and almost as good as using the target F0 contour",

author = "Ben Gillett and Simon King",

note = "ISSN: 1990-9772",

year = "2003",

month = sep,

day = "1",

language = "English",

isbn = "ISSN: 1990-9772",

pages = "101--104",

booktitle = "Eurospeech 2003 - Interspeech 2003",

publisher = "International Speech Communication Association",

}

TY - GEN

T1 - Transforming F0 Contours

AU - Gillett, Ben

AU - King, Simon

N1 - ISSN: 1990-9772

PY - 2003/9/1

Y1 - 2003/9/1

N2 - Voice transformation is the process of transforming the characteristics of speech uttered by a source speaker, such that a listener would believe the speech was uttered by a target speaker. Training F0 contour generation models for speech synthesis requires a large corpus of speech. If it were possible to adapt the F0 contour of one speaker to sound like that of another speaker, using a small, easily obtainable parameter set, this would be extremely valuable. We present a new method for the transformation of F0 contours from one speaker to another based on a small linguistically motivated parameter set. The system performs a piecewise linear mapping using these parameters. A perceptual experiment clearly demonstrates that the presented system is at least as good as an existing technique for all speaker pairs, and that in many cases it is much better and almost as good as using the target F0 contour

AB - Voice transformation is the process of transforming the characteristics of speech uttered by a source speaker, such that a listener would believe the speech was uttered by a target speaker. Training F0 contour generation models for speech synthesis requires a large corpus of speech. If it were possible to adapt the F0 contour of one speaker to sound like that of another speaker, using a small, easily obtainable parameter set, this would be extremely valuable. We present a new method for the transformation of F0 contours from one speaker to another based on a small linguistically motivated parameter set. The system performs a piecewise linear mapping using these parameters. A perceptual experiment clearly demonstrates that the presented system is at least as good as an existing technique for all speaker pairs, and that in many cases it is much better and almost as good as using the target F0 contour

M3 - Conference contribution

SN - ISSN: 1990-9772

SP - 101

EP - 104

BT - Eurospeech 2003 - Interspeech 2003

PB - International Speech Communication Association

ER -

Transforming F0 Contours

Abstract

Access to Document

Fingerprint

Projects

Concatenative Speech synthesis using articulatory information.

Cite this