# US 2026/0222495 A1 · Hybrid digital signal processing-artificial intelligence acoustic echo cancellation for virtual conferences

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0222495 A1
- Document ID: US20260222495A1
- Status: Published application
- Kind: A1
- Application number: 19/575,197
- Filed: 2026-03-23
- Published: 2026-07-30
- Earliest claimed: 2023-01-20
- Assignees: Zoom Communications, Inc.
- Applicants: Zoom Communications, Inc.
- Inventors: Jiachuan Deng; Cheng Lun Hu; Zhaofeng Jia; Qiyong Liu; Wei Wang; Yueguan Wang
- Web page: https://patentgrep.com/patent/US20260222495A1

## CPC

- H04M 9/082: {using echo cancellers (echo cancellers per se H04B3/23)}
- G10K 11/17823: {Reference signals, e.g. ambient acoustic environment}
- G10K 11/17854: {the filter being an adaptive filter}
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)
- G10K: SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR

## Abstract

Methods and systems provide hybrid DSP-AI acoustic echo cancellation for virtual conferences. A linear acoustic echo cancelation (AEC) can be performed on an input audio signal to filter out linear echo present in the input audio signal and generate a first filtered audio signal. A level of nonlinear echo present in the first filtered audio signal can then be determined. When the level of nonlinear echo satisfies a threshold, an artificial intelligence (AI)-based nonlinear AEC can be performed on the first filtered audio signal to generate an AI-filtered audio signal. When the level of nonlinear echo does not satisfy the threshold, a digital signal processing (DSP)-based nonlinear AEC can be performed on the first filtered audio signal to generate a second filtered audio signal.

## Claims

### Claim 1

A method comprising:
receiving an input audio signal from a microphone;
performing linear acoustic echo cancelation (AEC) on the input audio signal to obtain a first filtered audio signal;
determining one or more correlations between the input audio signal, the first filtered audio signal, and a reference signal;
determining that a level of nonlinear echo satisfies a threshold based on the one or more correlations;
executing a trained artificial intelligence (AI) model on the first filtered audio signal to perform nonlinear AEC and obtain a second filtered audio signal; and
outputting the second filtered audio signal.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving the reference signal from a remote microphone;
determining a first correlation between the input audio signal and the reference signal; and
in response to determining that the first correlation satisfies a first threshold, determining a second correlation between the first filtered audio signal and the reference signal.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
determining a difference between the first correlation and the second correlation; and
in response to determining that the difference satisfies a second threshold, determining that the level of nonlinear echo satisfies the threshold.

### Claim 4 (depends on 2)

The method of claim 2, further comprising:
in response to determining that the second correlation satisfies a third threshold, determining that the level of nonlinear echo satisfies the threshold.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
determining the one or more correlations between the input audio signal, the first filtered audio signal, and the reference signal based on a Pearson correlation formula.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
responsive to determining that the level of nonlinear echo satisfies the threshold, determining a type of environment where the microphone is located comprising analyzing the nonlinear echo present in the first filtered audio signal; and
selecting the trained AI model, from multiple available AI models, based on the type of environment where the microphone is located, wherein the multiple available AI models are trained with nonlinear echo data collected from multiple types of environment correspondingly.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
receiving a second input audio signal from the microphone;
receiving a second reference signal from a remote microphone;
performing the linear AEC on the second input audio signal using a first digital signal processing (DSP) algorithm to obtain a third filtered audio signal;
determining a first correlation between the second input audio signal and a second reference signal;
in response to determining that the first correlation does not satisfy a first threshold, determining that a second level of nonlinear echo does not satisfy the threshold; and
performing a nonlinear AEC on the third filtered audio signal using a second DSP algorithm to obtain a fourth filtered audio signal.

### Claim 8

A system comprising:
a communications interface;
a non-transitory computer-readable medium; and
one or more processors communicatively coupled to the communications interface and the non-transitory computer-readable medium, the one or more processors configured to execute processor-executable instructions stored in the non-transitory computer-readable medium to:
receive an input audio signal from a microphone;
perform linear acoustic echo cancelation (AEC) on the input audio signal to obtain a first filtered audio signal;
determine one or more correlations between the input audio signal, the first filtered audio signal, and a reference signal;
determine that a level of nonlinear echo satisfies a threshold based on the one or more correlations;
execute a trained artificial intelligence (AI) model on the first filtered audio signal to perform nonlinear AEC and obtain a second filtered audio signal; and
output the second filtered audio signal.

### Claim 9 (depends on 8)

The system of claim 8, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
receive the reference signal from a remote microphone;
determine a first correlation between the input audio signal and the reference signal; and
in response to determining that the first correlation satisfies a first threshold, determine a second correlation between the first filtered audio signal and the reference signal.

### Claim 10 (depends on 9)

The system of claim 9, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
determine a difference between the first correlation and the second correlation; and
in response to determining that the difference satisfies a second threshold, determine that the level of nonlinear echo satisfies the threshold.

### Claim 11 (depends on 9)

The system of claim 9, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
in response to determining that the second correlation satisfies a third threshold, determine that the level of nonlinear echo satisfies the threshold.

### Claim 12 (depends on 8)

The system of claim 8, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
determine the one or more correlations between the input audio signal, the first filtered audio signal, and the reference signal based on a Pearson correlation formula.

### Claim 13 (depends on 8)

The system of claim 8, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
responsive to determining that the level of nonlinear echo satisfies the threshold, determine a type of environment where the microphone is located comprising analyzing the nonlinear echo present in the first filtered audio signal; and
select the trained AI model, from multiple available AI models, based on the type of environment where the microphone is located, wherein the multiple available AI models are trained with nonlinear echo data collected from multiple types of environment correspondingly.

### Claim 14 (depends on 8)

The system of claim 8, wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
receive a second input audio signal from the microphone;
receive a second reference signal from a remote microphone;
perform the linear AEC on the second input audio signal using a first digital signal processing (DSP) algorithm to obtain a third filtered audio signal;
determine a first correlation between the second input audio signal and a second reference signal;
in response to determining that the first correlation does not satisfy a first threshold, determine that a second level of nonlinear echo does not satisfy the threshold; and
perform a nonlinear AEC on the third filtered audio signal using a second DSP algorithm to obtain a fourth filtered audio signal.

### Claim 15

A non-transitory computer-readable medium comprising processor-executable instructions configured to cause one or more processors to:
receive an input audio signal from a microphone;
perform linear acoustic echo cancelation (AEC) on the input audio signal to obtain a first filtered audio signal;
determine one or more correlations between the input audio signal, the first filtered audio signal, and a reference signal;
determine that a level of nonlinear echo satisfies a threshold based on the one or more correlations;
execute a trained artificial intelligence (AI) model on the first filtered audio signal to perform nonlinear AEC and obtain a second filtered audio signal; and
output the second filtered audio signal.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, further comprising processor-executable instructions configured to cause one or more processors to:
receive the reference signal from a remote microphone;
determine a first correlation between the input audio signal and the reference signal; and
in response to determining that the first correlation satisfies a first threshold, determine a second correlation between the first filtered audio signal and the reference signal.

### Claim 17 (depends on 16)

The non-transitory computer-readable medium of claim 16, further comprising processor-executable instructions configured to cause one or more processors to:
determine a difference between the first correlation and the second correlation; and
in response to determining that the difference satisfies a second threshold, determine that the level of nonlinear echo satisfies the threshold.

### Claim 18 (depends on 16)

The non-transitory computer-readable medium of claim 16, further comprising processor-executable instructions configured to cause one or more processors to:
in response to determining that the second correlation satisfies a third threshold, determine that the level of nonlinear echo satisfies the threshold.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, further comprising processor-executable instructions configured to cause one or more processors to:
determine the one or more correlations between the input audio signal, the first filtered audio signal, and the reference signal based on a Pearson correlation formula.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium of claim 15, further comprising processor-executable instructions configured to cause one or more processors to:
receive a second input audio signal from the microphone;
receive a second reference signal from a remote microphone;
perform the linear AEC on the second input audio signal using a first digital signal processing (DSP) algorithm to obtain a third filtered audio signal;
determine a first correlation between the second input audio signal and a second reference signal;
in response to determining that the first correlation does not satisfy a first threshold, determine that a second level of nonlinear echo does not satisfy the threshold; and
perform a nonlinear AEC on the third filtered audio signal using a second DSP algorithm to obtain a fourth filtered audio signal.
