IP Library Granted Patent US 12,424,239
Granted Patent B2
US 12,424,239 · App. 17/942,289 · Granted Sep 23, 2025

System and method for acoustic channel identification-based data verification

Inventors: Dushyant Sharma (Tracy, CA); Uwe Helmut Jost (Groton, MA); Patrick Aubrey Naylor (Reading, GB); Ljubomir Milanovic (Vienna, AT); William Francis Ganong, III (Brookline, MA)
Assignee: Microsoft Technology Licensing, LLC.
G10L25/51G10L25/87H04S7/302H04S2420/01
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 12,424,239
App. No.
17/942,289
Granted
Sep 23, 2025
Kind
B2
Abstract

A method, computer program product, and computing system for determining a plurality of transfer functions for a plurality of corresponding segments from a reference recording and a suspect recording. A delta transfer function between the plurality of transfer functions of a pair of corresponding segments of the plurality of corresponding segments is determined. A recording comparison confidence score is generated for the pair of corresponding segments based upon, at least in part, the delta transfer function. The suspect recording is verified based upon, at least in part, the plurality of recording comparison confidence scores.

Claims (44)

1. A computer-implemented method, executed on a computing device, comprising:

aligning a reference recording with a target recording;

determining a plurality of transfer functions for a plurality of corresponding segments from the reference recording and the target recording;

determining a delta transfer function between a subset of transfer functions within the plurality of transfer functions, the subset of transfer functions corresponding to a pair of corresponding segments in the plurality of corresponding segments;

generating a recording comparison confidence score based upon, at least in part, the delta transfer function; and

verifying the target recording based upon, at least in part, the recording comparison confidence score.

2. The computer-implemented method of claim 1 , wherein each of the reference recording and the target recording is a video recording or an audio recording.

3. The computer-implemented method of claim 1 , wherein aligning the reference recording with the target recording includes:

identifying a plurality of voice active segments within each of the reference recording and the target recording.

4. The computer-implemented method of claim 3 , wherein identifying the plurality of voice active segments within each of the reference recording and the target recording includes:

using video information to identify a plurality of segments with minimal speaker movement.

5. The computer-implemented method of claim 3 , wherein determining the plurality of transfer functions for the plurality of segments from the reference recording and the target recording includes:

determining the plurality of transfer functions for the plurality of voice active segments.

6. The computer-implemented method of claim 1 , further comprising:

generating an acoustic signature of the target recording; and

identifying a recording source of the target recording based upon, at least in part, the acoustic signature of the target recording to an acoustic signature of the reference recording.

7. The computer-implemented method of claim 1 , wherein verifying the target recording includes:

aggregating a plurality of recording comparison confidence scores to generate an aggregated recording comparison confidence score;

determining whether the aggregated recording comparison confidence score is greater than a predefined threshold; and

in response to determining that the aggregated recording comparison confidence score is greater than the predefined threshold, verifying the target recording.

8. The computer-implemented method of claim 1 , wherein verifying the target recording includes at least one of:

determining whether the target recording is an authorized version of the reference recording based upon, at least in part, the recording comparison confidence score; and

identifying a recording source of the target recording based upon, at least in part, the recording comparison confidence score.

9. A computing system comprising:

a memory storing programming instructions; and

a processor configured to execute the programming instructions, wherein the programing instructions, upon execution by the processor, cause the computing system to align a suspect recording with a recording associated with a known acoustic environment, to determine a plurality of acoustic transfer functions for a plurality of corresponding segments from the suspect recording and the recording associated with the known acoustic environment, to generate an acoustic signature of the suspect recording and an acoustic signature of the recording associated with the known acoustic environment, and to verify the suspect recording based upon, at least in part, the acoustic signature of the suspect recording and the acoustic signature of the recording associated with the known acoustic environment.

10. The computing system of claim 9 , wherein aligning the suspect recording with the recording associated with the known acoustic environment includes:

identifying a plurality of voice active segments within each recording.

11. The computing system of claim 10 , wherein determining the plurality of transfer functions for a plurality of corresponding segments includes:

determining the plurality of transfer functions for the plurality of voice active segments.

12. The computing system of claim 9 , wherein determining the plurality of transfer functions for the plurality of corresponding segments includes:

using video information to identify a plurality of segments with minimal speaker movement.

13. The computing system of claim 9 , wherein the programming instructions further cause the computing system to identify a recording source of the suspect recording based upon, at least in part, comparing the acoustic signature of the suspect recording to the acoustic signature of the recording associated with the known acoustic environment.

14. A computer program product residing on a non-transitory computer readable medium having a plurality of instructions stored thereon which, when executed by a processor, cause the processor to perform operations comprising:

determining a plurality of acoustic transfer functions for a plurality of segments from a suspect recording, wherein determining the plurality of acoustic transfer functions for the plurality of segments from the suspect recording includes identifying a plurality of voice active segments within the suspect recording;

identifying a change over time in the plurality of acoustic transfer functions for the plurality of segments; and

in response to identifying the change over time in the plurality of acoustic transfer functions, determining that the suspect recording is a tampered recording.

15. The computer program product of claim 14 , wherein identifying the plurality of voice active segments within the suspect recording includes:

using video information to identify a plurality of segments with minimal speaker movement.

16. The computer program product of claim 14 , wherein determining the plurality of transfer functions for a plurality of segments from the suspect recording includes:

determining the plurality of transfer functions for the plurality of voice active segments.

17. The computer program product of claim 14 , wherein the operations further comprise:

generating a plurality of acoustic signatures for the plurality of segments from the suspect recording; and

identifying a change over time in the plurality of acoustic signatures for the plurality of segments from the suspect recording.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 6, 2025
From: NUANCE COMMUNICATIONS, INC.
To: MICROSOFT TECHNOLOGY LICENSING, LLC
Reel/Frame 070746/0729 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 14, 2023
From: NUANCE COMMUNICATIONS, INC.
To: MICROSOFT TECHNOLOGY LICENSING, LLC
Reel/Frame 065578/0676 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 12, 2022
From: SHARMA, DUSHYANT; JOST, UWE HELMUT; NAYLOR, PATRICK AUBREY; MILANOVIC, LJUBOMIR; GANONG, WILLIAM FRANCIS, III
To: NUANCE COMMUNICATIONS, INC.
Reel/Frame 061058/0581 →