SPEECH RECOGNITION DEVICE, ACOUSTIC MODEL LEARNING DEVICE, SPEECH RECOGNITION METHOD, AND ACOUSTIC MODEL LEARNING METHOD

Title:

SPEECH RECOGNITION DEVICE, ACOUSTIC MODEL LEARNING DEVICE, SPEECH RECOGNITION METHOD, AND ACOUSTIC MODEL LEARNING METHOD

Document Type and Number:

WIPO Patent Application WO/2016/129110

Kind Code:

A1

Abstract:

A speech recognition device comprises: a feature-space transformation matrix estimation unit (6) for estimating a plurality of feature-space transformation matrices corresponding to a regression tree via linear regression for feature space using a posterior probability for a first recognition result acquired by speech recognition, a regression tree created from an acoustic model, and a speech feature extracted from an input speech; and a feature conversion unit (7) for converting a speech feature using a feature-space transformation matrix determined on the basis of the acoustic model and the state series in the acoustic model from among the plurality of feature-space transformation matrices. A decoding unit (2) compares the converted speech feature with the acoustic model to perform speech recognition and thereby obtain a recognition result.

More Like This:

JP2007256297	SPEECH PROCESSING METHOD AND COMMUNICATION SYSTEM, AND COMMUNICATION TERMINAL AND SERVER AND PROGRAM
WO/2006/033044	METHOD OF TRAINING A ROBUST SPEAKER-DEPENDENT SPEECH RECOGNITION SYSTEM WITH SPEAKER-DEPENDENT EXPRESSIONS AND ROBUST SPEAKER-DEPENDENT SPEECH RECOGNITION SYSTEM
JPS59180596	SPEAKER ADAPTED VOICE RECOGNITION SYSTEM

Inventors:

KANAGAWA HIROKI (JP)
TACHIOKA YUKI (JP)

Application Number:

PCT/JP2015/053998

Publication Date:

August 18, 2016

Filing Date:

February 13, 2015

Export Citation:

Click for automatic bibliography generation Help

Assignee:

MITSUBISHI ELECTRIC CORP (JP)

International Classes:

G10L15/07

Foreign References:

JP2013178343A

2013-09-09

Other References:

XIN LEI ET AL.: "Robust feature space adaptation for telephony speech recognition", PROC. INTERSPEECH 2006, 17 September 2006 (2006-09-17), pages 773 - 776
M.J.F. GALES: "Maximum Likelihood Linear Transformations for HMM-Based Speech Recognition", COMPUTER SPEECH AND LANGUAGE, May 1997 (1997-05-01), pages 1 - 19

Attorney, Agent or Firm:

TAZAWA, Hideaki et al. (JP)
Hideaki Tazawa (JP)

Download PDF:

View/Download PDF PDF Help

Previous Patent: SEMICONDUCTOR DEVICE AND METHOD FOR MANUFACTURING SAME

Next Patent: MEMBRANE DEVICE AND METHOD FOR MANUFACTURING SAME