IP Library Granted Patent US 8,073,689
Granted Patent B2
US 8,073,689 · App. 11/331,806 · Granted Dec 6, 2011

Repetitive transient noise removal

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,073,689
App. No.
11/331,806
Granted
Dec 6, 2011
Kind
B2
Abstract

A system improves the perceptual quality of a speech signal by dampening undesired repetitive transient noises. The system includes a repetitive transient noise detector adapted to detect repetitive transient noise in a received signal. The received signal may include a harmonic and a noise spectrum. The system further includes a repetitive transient noise attenuator that substantially removes or dampens repetitive transient noises from the received signal. The method of dampening the repetitive transient noises includes modeling characteristics of repetitive transient noises; detecting characteristics in the received signal that correspond to the modeled characteristics of the repetitive transient noises; and substantially removing components of the repetitive transient noises from the received signal that correspond to some or all of the modeled characteristics of the repetitive transient noises.

Claims (58)

1. A system for suppressing repetitive transient noises from a signal comprising:

a repetitive transient noise detector that comprises a processor adapted to detect the presence of transient noise in a received signal comprising a harmonic spectrum and a noise spectrum, where the repetitive transient noise detector is adapted to analyze one or more frequency spectrum sound event characteristics and one or more temporal sound event characteristics of the transient noise to determine whether the transient noise is repetitive transient noise; and

a repetitive transient noise attenuator adapted to dampen the transient noise detected in the received signal in response to the repetitive transient noise detector identifying the transient noise as being repetitive transient noise;

where the repetitive transient noise detector is adapted to fit a function to a selected portion of the received signal in a time-frequency domain to evaluate spectro-temporal shape characteristics of a sound event in the received signal and to determine whether the transient noise is repetitive transient noise; and

where the repetitive transient noise detector is adapted to identify the sound event as a repetitive transient noise event based on the function, a signal envelope of the sound event, and a correlation threshold that depends on a width and sharpness of the repetitive transient noise.

2. The system of claim 1 where the repetitive transient noise detector comprises a model of repetitive transient noise and the repetitive transient noise detector is adapted to compare an attribute of the received signal with an attribute of the model of the repetitive transient noise.

3. The system of claim 2 where the model of the repetitive transient noise comprises an average repetitive transient noise model based on multiple repetitive transient noise models.

4. The system of claim 2 where the model comprises a spectral component and a temporal component, and where the temporal component comprises a first sound event and a second substantially similar sound event separated in time.

5. The system of claim 4 where a period of time between the first sound event and the second sound event is estimated through an adaptive model.

6. The system of claim 2 where the model comprises a spectral component and a temporal component, and where the spectral component comprises one or more attributes of a spectral shape of a sound event associated with a repetitive transient noise.

7. The system of claim 6 where the attributes of the spectral shape of a sound event associated with a repetitive transient noise comprises a broadband frequency response.

8. The system of claim 7 further comprising a vehicle that transports the repetitive transient noise detector and the repetitive transient noise attenuator.

9. The system of claim 1 where the repetitive transient noise detector is adapted to estimate or measure a temporal spacing between multiple sound events of the transient noise and compare the temporal spacing to a modeled temporal spacing of a model repetitive transient noise to identify whether the transient noise is repetitive.

10. The system of claim 1 where the repetitive transient noise detector is adapted to identify a first sound event in the received signal as being a potential repetitive transient noise, where the repetitive transient noise detector is adapted to analyze one or more portions of the received signal before or after the first sound event to determine whether the received signal contains a second sound event with the same or similar characteristics as the first sound event, and where the repetitive transient noise detector is adapted to determine that the received signal comprises repetitive transient noise upon locating the second sound event.

11. The system of claim 1 further comprising a background noise estimator and a transient detector, where the transient detector disables the background noise estimator in response to a determination that an instantaneous background noise level exceeds an average background noise level by more than a pre-determined amount.

12. The system of claim 1 where in response to identifying a first sound event as a potential repetitive transient noise, the repetitive transient noise detector looks back to previously analyzed time windows to determine whether a corresponding sound event of a repetitive transient noise was received, and where the repetitive transient noise detector identifies the first sound as a repetitive transient noise in response to locating the corresponding sound event.

13. The system of claim 1 where in response to identifying a first sound event as a potential repetitive transient noise, the repetitive transient noise detector looks forward to later received time windows to determine whether a corresponding sound event of a repetitive transient noise will be received, and where the repetitive transient noise detector identifies the first sound as a repetitive transient noise in response to locating the corresponding sound event.

14. The system of claim 1 where the repetitive transient noise detector is adapted to determine a probability that the received signal includes repetitive transient noise, and identify a sound event as repetitive transient noise when the probability exceeds a predetermined threshold.

15. A repetitive transient noise detector for detecting the presence of repetitive noise in a signal, the repetitive transient noise detector comprising:

an analog to digital converter for converting a received signal into a digital signal;

a windowing function generator for dividing the received signal into a plurality of individual analysis windows;

a transform module for transforming the individual analysis windows from a time domain spectra to a frequency domain spectra;

a modeler that generates and stores attributes of repetitive transient noise in a memory; and

a controller adapted to fit a function to a selected portion of the received signal in a time-frequency domain to evaluate spectro-temporal shape characteristics of a transient noise in the received signal and to determine whether the transient noise is repetitive transient noise, where the controller is adapted to identify the transient noise as a repetitive transient noise event based on the function, a signal envelope of the sound event, and a correlation threshold that depends on a width and sharpness of the repetitive transient noise;

where the modeler is adapted to identify the transient noise for attenuation in response to the modeler determining that the transient noise is repetitive transient noise.

16. The repetitive transient noise detector of claim 15 where the model attributes comprise average model attributes based on attributes of multiple models.

17. The repetitive transient noise detector of claim 15 where the windowing function generator comprises a Hanning window function generator.

18. The repetitive transient noise detector of claim 15 where the transform module performs a Fast Fourier Transform on the plurality of individual analysis windows.

19. The repetitive transient noise detector of claim 15 where the model attributes comprise temporal characteristics of repetitive transient noises.

20. The repetitive transient noise detector of claim 19 where the model attributes comprise spectral characteristics of repetitive transient noises.

21. The repetitive transient noise detector of claim 15 where the model attributes comprise temporal characteristics and spectral characteristics of estimated repetitive transient noises.

22. The repetitive transient noise detector of claim 21 where the model attributes represent a plurality of sound events having substantially similar spectral characteristics separated by a short time period.

23. The repetitive transient noise detector of claim 22 where the model attributes comprise spectral shape characteristics of the plurality of sound events.

24. The repetitive transient noise detector of claim 15 further comprising a residual attenuator for tracking the power spectrum of the received signal.

25. The repetitive transient noise detector of claim 15 where the controller is adapted to analyze attributes of the spectra of the transformed analysis windows to determine whether the transient noise is repetitive transient noise.

26. The repetitive transient noise detector of claim 15 where the controller is adapted to estimate or measure a temporal spacing between multiple sound events of the transient noise and compare the temporal spacing to a modeled temporal spacing of a model repetitive transient noise to identify whether the transient noise is repetitive.

27. The repetitive transient noise detector of claim 15 where the modeler comprises a non-transitory computer-readable medium or circuit.

28. A method of attenuating repetitive transient noises from a signal comprising:

fitting a function to a received signal in a time-frequency domain to evaluate spectro-temporal shape characteristics of a transient noise in the received signal and to determine whether the transient noise is repetitive transient noise;

identifying the transient noise as a repetitive transient noise event based on the function, a signal envelope of the transient noise, and a correlation threshold that depends on a width and sharpness of the repetitive transient noise; and

attenuating, by a processor, at least a portion of the transient noise from the received signal in response to identifying the transient noise as the repetitive transient noise event.

29. The method of claim 28 further comprising:

deriving an average repetitive transient noise model from multiple modeled characteristics of repetitive transient noises;

determining whether the characteristics of the signal correspond to characteristics of the average repetitive transient noise model.

30. The method of claim 28 further comprising:

identifying a first sound event in the signal as being a potential repetitive transient noise;

analyzing one or more portions of the signal before or after the first sound event to determine whether the signal contains a second sound event with the same or similar characteristics as the first sound event; and

determining that the signal comprises repetitive transient noise upon locating the second sound event.

31. The method of claim 28 further comprising modeling spectral shape attributes of repetitive transient noises.

32. The method of claim 31 where the spectral shape attributes of the sound events occur across a broadband frequency.

33. A system for suppressing repetitive transient noises from a signal comprising:

a repetitive transient noise detector that comprises a processor adapted to fit a function to a received signal in a time-frequency domain to evaluate spectro-temporal shape characteristics of a transient noise in the received signal and to determine whether the transient noise is repetitive transient noise; and

a repetitive transient noise attenuator adapted to dampen the transient noise in the received signal in response to the repetitive transient noise detector identifying the transient noise as being repetitive;

where the repetitive transient noise detector is adapted to identify the transient noise as a repetitive transient noise event based on the function, a signal envelope of the transient noise, and a correlation threshold that depends on a width and sharpness of the repetitive transient noise.

34. A system for suppressing repetitive transient noises from a signal comprising:

a repetitive transient noise detector that comprises a processor adapted to detect the presence of transient noise in a received signal comprising a harmonic spectrum and a noise spectrum, where the repetitive transient noise detector is adapted to fit a function to a selected portion of the received signal in a time-frequency domain to evaluate spectro-temporal shape characteristics of a sound event in the received signal and to determine whether the transient noise is repetitive transient noise, where the repetitive transient noise detector is adapted to identify the sound event as a repetitive transient noise event based on a correlation between the function and a signal envelope of the sound event; and

a repetitive transient noise attenuator adapted to dampen the transient noise detected in the received signal in response to the repetitive transient noise detector identifying the transient noise as being repetitive transient noise;

where the repetitive transient noise detector is adapted to identify the sound event as the repetitive transient noise event based on the function, the signal envelope of the sound event, and a correlation threshold that depends on a width and sharpness of the repetitive transient noise.

Assignments (15)
CORRECTIVE ASSIGNMENT TO CORRECT THE ADDED PATENT NUMBER TO REMOVE PATENT NO. 8,873,407 AT PREVIOUSLY RECORDED ON REEL 64066 FRAME 1. ASSIGNOR(S) HEREBY CONFIRMS THE NUNC PRO TUNC ASSIGNMENT EFFECTIVE DATE MARCH 20, 2023. Recorded Feb 2, 2026
From: BLACKBERRY LIMITED
To: MALIKIE INNOVATIONS LIMITED
Reel/Frame 074921/0001 →
CORRECTIVE ASSIGNMENT TO CORRECT 12817157 APPLICATION NUMBER PREVIOUSLY RECORDED AT REEL: 064015 FRAME: 0001. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Sep 5, 2023
From: OT PATENT ESCROW, LLC
To: MALIKIE INNOVATIONS LIMITED
Reel/Frame 064807/0001 →
CORRECTIVE ASSIGNMENT TO CORRECT THE COVER SHEET AT PAGE 50 TO REMOVE 12817157 PREVIOUSLY RECORDED ON REEL 063471 FRAME 0474. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Sep 5, 2023
From: BLACKBERRY LIMITED
To: OT PATENT ESCROW, LLC
Reel/Frame 064806/0669 →
NUNC PRO TUNC ASSIGNMENT Recorded Jun 19, 2023
From: BLACKBERRY LIMITED
To: MALIKIE INNOVATIONS LIMITED
Reel/Frame 064066/0001 →
NUNC PRO TUNC ASSIGNMENT Recorded Jun 16, 2023
From: OT PATENT ESCROW, LLC
To: MALIKIE INNOVATIONS LIMITED
Reel/Frame 064015/0001 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 27, 2023
From: BLACKBERRY LIMITED
To: OT PATENT ESCROW, LLC
Reel/Frame 063471/0474 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 22, 2020
From: 2236008 ONTARIO INC.
To: BLACKBERRY LIMITED
Reel/Frame 053313/0315 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 4, 2014
From: 8758271 CANADA INC.
To: 2236008 ONTARIO INC.
Reel/Frame 032607/0674 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 4, 2014
From: QNX SOFTWARE SYSTEMS LIMITED
To: 8758271 CANADA INC.
Reel/Frame 032607/0943 →
CHANGE OF NAME Recorded Feb 27, 2012
From: QNX SOFTWARE SYSTEMS CO.
To: QNX SOFTWARE SYSTEMS LIMITED
Reel/Frame 027768/0863 →
CONFIRMATORY ASSIGNMENT Recorded Jul 9, 2010
From: QNX SOFTWARE SYSTEMS (WAVEMAKERS), INC.
To: QNX SOFTWARE SYSTEMS CO.
Reel/Frame 024659/0370 →
PARTIAL RELEASE OF SECURITY INTEREST Recorded Jun 3, 2010
From: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT
To: HARMAN INTERNATIONAL INDUSTRIES, INCORPORATED; QNX SOFTWARE SYSTEMS (WAVEMAKERS), INC.; QNX SOFTWARE SYSTEMS GMBH & CO. KG
Reel/Frame 024483/0045 →
SECURITY AGREEMENT Recorded May 8, 2009
From: HARMAN INTERNATIONAL INDUSTRIES, INCORPORATED; BECKER SERVICE-UND VERWALTUNG GMBH; CROWN AUDIO, INC.; HARMAN BECKER AUTOMOTIVE SYSTEMS (MICHIGAN), INC.; HARMAN BECKER AUTOMOTIVE SYSTEMS HOLDING GMBH; HARMAN BECKER AUTOMOTIVE SYSTEMS, INC.; HARMAN CONSUMER GROUP, INC.; HARMAN DEUTSCHLAND GMBH; HARMAN FINANCIAL GROUP LLC; HARMAN HOLDING GMBH & CO. KG; HARMAN MUSIC GROUP, INCORPORATED; HARMAN SOFTWARE TECHNOLOGY INTERNATIONAL BETEILIGUNGS GMBH; HARMAN SOFTWARE TECHNOLOGY MANAGEMENT GMBH; HBAS INTERNATIONAL GMBH; HBAS MANUFACTURING, INC.; INNOVATIVE SYSTEMS GMBH NAVIGATION-MULTIMEDIA; JBL INCORPORATED; LEXICON, INCORPORATED; MARGI SYSTEMS, INC.; QNX SOFTWARE SYSTEMS (WAVEMAKERS), INC.; QNX SOFTWARE SYSTEMS CANADA CORPORATION; QNX SOFTWARE SYSTEMS CO.; QNX SOFTWARE SYSTEMS GMBH; QNX SOFTWARE SYSTEMS GMBH & CO. KG; QNX SOFTWARE SYSTEMS INTERNATIONAL CORPORATION; QNX SOFTWARE SYSTEMS, INC.; XS EMBEDDED GMBH (F/K/A HARMAN BECKER MEDIA DRIVE TECHNOLOGY GMBH)
To: JPMORGAN CHASE BANK, N.A.
Reel/Frame 022659/0743 →
CHANGE OF NAME Recorded Nov 14, 2006
From: HARMAN BECKER AUTOMOTIVE SYSTEMS - WAVEMAKERS, INC.
To: QNX SOFTWARE SYSTEMS (WAVEMAKERS), INC.
Reel/Frame 018515/0376 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 13, 2006
From: HETHERINGTON, PHILLIP A.; PARANJPE, SHREYAS A.
To: HARMAN BECKER AUTOMOTIVE SYSTEMS - WAVEMAKERS, INC.
Reel/Frame 018512/0080 →