DISCRIMINATIVE TRAINING OF AUTOMATIC SPEECH RECOGNITION MODELS WITH NATURAL LANGUAGE PROCESSING DICTIONARY FOR SPOKEN LANGUAGE PROCESSING

Patent №

US 10,140,976

Granted

2018-11-27

Filed 2015

Owner

INTERNATIONAL BUSINESS MACHINES CORPORATION

AI components

4

ml · nlp · speech · kr

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

14968439

Methods and systems for language processing includes training one or more automatic speech recognition models using an automatic speech recognition dictionary. A set of N automatic speech recognition hypotheses for an input is determined, based on the one or more automatic speech recognition models, using a processor. A best hypothesis is selected using a discriminative language model and a list of relevant words. Natural language processing is performed on the best hypothesis.

Machine learningNatural languageSpeechKnowledge representationG10L 15/063G06F 40/242G06F 40/284G10L 15/01G10L 15/183G10L 15/197G10L 15/26G10L 2015/0635+1 more

AI classification

Natural language1.00
Speech1.00
Machine learning1.00
Knowledge representation0.85
AI hardware0.23
Evolutionary computation0.04
Planning0.00
Vision0.00

Ownership

INTERNATIONAL BUSINESS MACHINES CORPORATION

assignment · 373170672

Assignors

SUZUKI, MASAYUKI, ITOH, NOBUYASU, KURATA, GAKUTO, NAGANO, TOHRU, SUZUKI, MASAYUKI, ITOH, NOBUYASU, KURATA, GAKUTO, NAGANO, TOHRU

On an employer assignment, the assignors are typically the inventors.

From the same owner

© 2026 NYSGPT2525 LLC