(19)
(11) EP 1 860 646 A3

(12) EUROPEAN PATENT APPLICATION

(88) Date of publication A3:
03.09.2008 Bulletin 2008/36

(43) Date of publication A2:
28.11.2007 Bulletin 2007/48

(21) Application number: 07116266.3

(22) Date of filing: 27.03.2003
(51) International Patent Classification (IPC): 
G10L 13/06(2006.01)
G10L 15/14(2006.01)
(84) Designated Contracting States:
DE FI FR GB NL

(30) Priority: 29.03.2002 US 369043
14.01.2003 US 341869

(62) Application number of the earlier application in accordance with Art. 76 EPC:
03100795.8 / 1394769

(71) Applicant: AT&T Corp.
New York, NY 10013-2412 (US)

(72) Inventors:
  • Conkie, Alistair, D.
    Morris County, NJ 07960 (US)
  • Kim, Yeon-Jun
    Morris County, NJ 07981 (US)

(74) Representative: Suckling, Andrew Michael 
Marks & Clerk 4220 Nash Court
Oxford Business Park South Oxford Oxfordshire OX4 2RU
Oxford Business Park South Oxford Oxfordshire OX4 2RU (GB)

   


(54) Automatic segmentaion in speech synthesis


(57) A method for segmenting phone labels to reduce misalignments in order to improve synthetic speech when the phone labels are concatenated comprises:
training a set of HMMs using one of a specific speaker's hand-labeled speech data and speaker-independent speech data;
segmenting the trained set of HMMs using an alignment to produce phone labels, wherein each phone label has a spectral boundary;
using a weighted slope metric to identify bending points of spectral transitions, wherein each bending point corresponds to a spectral boundary; and
correcting a particular spectral boundary of a particular phone label if the particular spectral boundary does not coincide with a particular bending point.





Search report