DYNAMIC THRESHOLD METHODS FOR FILTERING NOISE AND RESTORING ATTENUATED HIGH-FREQUENCY COMPONENTS OF ACOUSTIC SIGNALS

Patent №

US 9,684,087

Granted

2017-06-20

Filed 2014

Owner

SAUDI ARABIAN OIL COMPANY

Lab

AI components

1

speech

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

14485562

Methods of providing for filtering noise and/or restoring attenuated spectral components in acoustic signals, are provided. An exemplary embodiment of a method includes dynamically filtering each of a plurality of raw FFT data samples of a record to remove or attenuate background noise contained therein to thereby produce a corresponding plurality of cleaned FFT data samples. The sample-specific background noise is removed or attenuated by a record-specific dynamic filter to produce the corresponding cleaned FFT data samples. The method can also include restoring the attenuated high-frequency components of the cleaned data samples through application of a record-specific Restoring Processor at least partially defined by a portion of the cleaned data samples and a Gain Function to thereby produce cleaned and restored data samples, and applying an inverse transformation to convert the cleaned and restored data samples into cleaned and restored data samples in time domain data.

SpeechG01V 1/364G01N 29/11G01N 29/32G10L 21/0208G01V 1/40G01V 2210/324G01V 2210/40G10L 21/0232+1 more

AI classification

Speech0.92
Machine learning0.44
AI hardware0.18
Knowledge representation0.17
Natural language0.00
Planning0.00
Vision0.00
Evolutionary computation0.00

Ownership

SAUDI ARABIAN OIL COMPANY

assignment · 339250393

Assignors

YANG, YUNLAI

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

From the same owner

© 2026 NYSGPT2525 LLC