Title:
話し手に暗黙的に順応する技術を用いた音声認識システム
Document Type and Number:
Japanese Patent JP2004530155
Kind Code:
A
Abstract:
A voice recognition (VR) system is disclosed that utilizes a combination of speaker independent (SI) and speaker dependent (SD) acoustic models. At least one SI acoustic model is used in combination with at least one SD acoustic model to provide a level of speech recognition performance that at least equals that of a purely SI acoustic model. The disclosed hybrid SI/SD VR system continually uses unsupervised training to update the acoustic templates in the one or more SD acoustic models. The hybrid VR system then uses the updated SD acoustic models in combination with the at least one SI acoustic model to provide improved VR performance during VR testing
Inventors:
Malayas, Narendranas
Dejaco, Andrew Pee
Chan, Chiang Chun
Jaril, suhail
Bi, nin
Gardadori, Halinas
Dejaco, Andrew Pee
Chan, Chiang Chun
Jaril, suhail
Bi, nin
Gardadori, Halinas
Application Number:
JP2002578283A
Publication Date:
September 30, 2004
Filing Date:
March 22, 2002
Export Citation:
Assignee:
QUALCOMM INCORPORATED
International Classes:
G10L15/10; G10L15/12; G10L15/14; G10L15/16; G10L17/04; G10L15/06; (IPC1-7): G10L15/06; G10L15/00; G10L15/10; G10L15/12; G10L15/14; G10L15/16
Attorney, Agent or Firm:
Takehiko Suzue
Satoshi Kono
Makoto Nakamura
Yoshihiro Fukuhara
Sadao Muramatsu
Ryo Hashimoto
Satoshi Kono
Makoto Nakamura
Yoshihiro Fukuhara
Sadao Muramatsu
Ryo Hashimoto