Presentation	1997/3/7 Pause estimation by network model for text-to-speech synthesis Takashi EBIHARA, Yasushi ISHIKAWA,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	In this paper, we describe a method of pause estimation for Japanese text-to-speech synthesis. We have proposed a network model which integrates prosodic control and syntactic analysis. In a network model, each state represents linguistic and prosodic state of sentence, and a prosodic parameter is generated when taking transition. We applied the proposed prosodic control model to pause estimation. In the model, one node reprenets pause, and every node has transition are to pause node with probability. At each node, pause occurring rate is calculated to decide the pause. The prediction results showed our model can predict pauses in a sentence with high accuracy.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	speech synthesis / pause / prosody / pitch frequency control
Paper #	SP96-133
Date of Issue

Conference Information
Committee	SP
Conference Date	1997/3/7(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Speech (SP)
Language	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Pause estimation by network model for text-to-speech synthesis
Sub Title (in English)
Keyword(1)	speech synthesis
Keyword(2)	pause
Keyword(3)	prosody
Keyword(4)	pitch frequency control
1st Author's Name	Takashi EBIHARA
1st Author's Affiliation	Information Technology RampD Center, MITSUBISHI Electric Corporation()
2nd Author's Name	Yasushi ISHIKAWA
2nd Author's Affiliation	Information Technology RampD Center, MITSUBISHI Electric Corporation
Date	1997/3/7
Paper #	SP96-133
Volume (vol)	vol.96
Number (no)	566
Page	pp.pp.-
#Pages	6
Date of Issue