SPEECH RECOGNITION METHOD USING TIME-FREQUENCY MASKING MECHANISM

Patent №

US 5,459,815

Granted

1995-10-17

Filed 1993

Owner

ATR AUDITORY AND VISUAL PERCEPTION RESEARCH LABORATORIES

Lab

AI components

4

ml · nlp · speech · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

08079425

A speech recognition method in which input speech signals are converted to digital signals and then time sequentially converted to cepstrum coefficients or logarithmic spectra. Dynamic spectrum time sequence is obtained by time frequency filtering of cepstrum coefficients, or masked spectrum time sequence is obtained by time frequency masking of the logarithmic vector time sequence. Based on the dynamic cepstrum time sequence or masked spectrum time sequence obtained in this manner, speech is recognized.

AI classification

Speech1.00
Natural language1.00
Machine learning0.90
AI hardware0.55
Vision0.03
Planning0.01
Evolutionary computation0.00
Knowledge representation0.00

Ownership

ATR AUDITORY AND VISUAL PERCEPTION RESEARCH LABORATORIES

assignment · 66010342

Assignors

AIKAWA, KIYOAKI, KAWAHARA, HIDEKI, TOHKURA, YOH'ICHI

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

© 2026 NYSGPT2525 LLC