Title:
キーワード検出装置、キーワード検出方法及びキーワード検出用コンピュータプログラム
Document Type and Number:
Japanese Patent JP6679898
Kind Code:
B2
Abstract:
A keyword detector includes a processor configured to calculate a feature vector for each frame from a speech signal, input the feature vector for each frame to a DNN to calculate a first output probability for each triphone according to a sequence of phonemes contained in a predetermined keyword and a second output probability for each monophone, for each of at least one state of an HMM, calculate a first likelihood representing the probability that the predetermined keyword is uttered in the speech signal by applying the first output probability to the HMM, calculate a second likelihood for the most probable phoneme string in the speech signal by applying the second output probability to the HMM, and determine whether the keyword is to be detected on the basis of the first likelihood and the second likelihood.
Inventors:
Shoji Hayakawa
Application Number:
JP2015228889A
Publication Date:
April 15, 2020
Filing Date:
November 24, 2015
Export Citation:
Assignee:
富士通株式会社
International Classes:
G10L15/10; G10L15/14; G10L15/16; G10L15/30
Domestic Patent References:
| JP2011185997A | ||||
| JP2015102806A |
Attorney, Agent or Firm:
Atsushi Aoki
Koichi Itsubo
Tsutomu Kono
Tetsuo Miyamoto
Koichi Itsubo
Tsutomu Kono
Tetsuo Miyamoto