# US 2025/0095666 A1 · Leveraging self-supervised speech representations for domain adaptation in speech enhancement

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

- Number: US 2025/0095666 A1
- Document ID: US20250095666A1
- Status: Published application
- Kind: A1
- Application number: 18/884,978
- Filed: 2024-09-13
- Published: 2025-03-20
- Earliest claimed: 2023-09-20
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Ching-Hua LEE; Chouchang YANG; Rakshith Sharma SRINIVASA; Yashas Malur SAIDUTTA; Jaejin CHO; Yilin SHEN; Hongxia JIN
- Web page: https://patentgrep.com/patent/US20250095666A1

## CPC

- G10L 21/0208: Noise filtering
- G10L 15/063: {Training}
- G10L 21/02: Speech enhancement, e.g. noise reduction or echo cancellation (reducing echo effects in line transmission systems H04B3/20; echo suppression in hands-free telephones H04M9/08)
- G10L 21/0216: characterised by the method used for estimating noise
- G10L 2021/02166: {Microphone arrays; Beamforming}
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method for generating a customized speech enhancement model includes obtaining noisy-clean speech data from a source domain, obtaining noisy speech data from a target domain; obtaining raw speech data, using the noisy-clean speech data, the noisy speech data, and the raw speech data, training the customized SE model based on at least one of self-supervised representation-based adaptation (SSRA), ensemble mapping, or self-supervised adaptation loss, generating the customized SE model by denoising the noisy speech data using the trained customized SE model, and providing the customized SE model to a user device to use the denoised noisy speech data.

## Claims

### Claim 1

A method for generating a customized speech enhancement (SE) model, performed by at least one processor of an electronic device, the method comprising:
obtaining noisy-clean speech data from a source domain;
obtaining noisy speech data from a target domain;
obtaining raw speech data;
using the noisy-clean speech data, the noisy speech data, and the raw speech data, training the customized SE model based on at least one of self-supervised representation-based adaptation (SSRA), ensemble mapping, or self-supervised adaptation loss;
generating the customized SE model by denoising the noisy speech data using the trained customized SE model; and
providing the customized SE model to a user device to use the denoised noisy speech data.

### Claim 2 (depends on 1)

The method of claim 1, wherein the training the customized SE model comprises training the customized SE model based on the SSRA, and the training the customized SE model further comprises pre-training a self-supervised learning (SSL) encoder in a self-supervised manner, providing a target domain enhanced signal to the SSL encoder, and providing source domain clean signals to the SSL encoder.

### Claim 3 (depends on 1)

The method of claim 1, wherein the training the customized SE model comprises training the customized SE model based on the ensemble mapping, and the training the customized SE model further comprises pseudo labeling the noisy speech data from the target domain.

### Claim 4 (depends on 1)

The method of claim 1, wherein the training the customized SE model comprises training the customized SE model based on the self-supervised adaptation loss, and the training the customized SE model further comprises using a distance metric in an SSRA loss term.

### Claim 5 (depends on 1)

The method of claim 1, wherein the noisy speech data is obtained from the user device in the target domain.

### Claim 6 (depends on 5)

The method of claim 5, wherein the user device comprises at least one of a mobile phone, a refrigerator, a smart watch, glasses, or a television.

### Claim 7 (depends on 1)

The method of claim 1, wherein the noisy speech data is obtained from a plurality of microphones corresponding to a plurality of user devices.

### Claim 8

A server device comprising:
a memory storing instructions; and
at least one processor,
wherein the instructions, when executed by the at least one processor, cause the server device to:
obtain noisy-clean speech data from a source domain;
obtain noisy speech data from a target domain;
obtain raw speech data;
using the noisy-clean speech data, the noisy speech data, and the raw speech data, train a customized SE model based on at least one of self-supervised representation-based adaptation (SSRA), ensemble mapping, or self-supervised adaptation loss;
generate the customized SE model by denoising the noisy speech data using the trained customized SE model; and
provide the customized SE model to a user device to use the denoised noisy speech data.

### Claim 9 (depends on 8)

The server device of claim 8, wherein the instructions, when executed by the at least one processor, cause the server device to pre-train a self-supervised learning (SSL) encoder in a self-supervised manner, provide a target domain enhanced signal to the SSL encoder, and provide source domain clean signals to the SSL encoder.

### Claim 10 (depends on 8)

The server device of claim 8, wherein the instructions, when executed by the at least one processor, cause the server device to train the customized SE model based on the ensemble mapping, and pseudo label the noisy speech data from the target domain.

### Claim 11 (depends on 8)

The server device of claim 8, wherein the instructions, when executed by the at least one processor, cause the server device to train the customized SE model based on the self-supervised adaptation loss, and use a distance metric in an SSRA loss term.

### Claim 12 (depends on 8)

The server device of claim 8, wherein the noisy speech data is obtained from the user device in the target domain.

### Claim 13 (depends on 12)

The server device of claim 12, wherein the user device comprises at least one of a mobile phone, a refrigerator, a smart watch, glasses, or a television.

### Claim 14 (depends on 8)

The server device of claim 8, wherein the noisy speech data is obtained from a plurality of microphones corresponding to a plurality of user devices.

### Claim 15

A non-transitory computer-readable recording medium configured to store instructions for generating a customized speech enhancement (SE) model, which, when executed by at least one processor of an electronic device, cause the at least one processor to perform a method comprising:
obtaining noisy-clean speech data from a source domain;
obtaining noisy speech data from a target domain;
obtaining raw speech data;
using the noisy-clean speech data, the noisy speech data, and the raw speech data, training the customized SE model based on at least one of self-supervised representation-based adaptation (SSRA), ensemble mapping, or self-supervised adaptation loss;
generating the customized SE model by denoising the noisy speech data using the trained customized SE model; and
providing the customized SE model to a user device to use the denoised noisy speech data.

### Claim 16 (depends on 15)

The non-transitory computer-readable recording medium of claim 15, wherein the training the customized SE model comprises training the customized SE model based on the SSRA, and the training the customized SE model further comprises pre-training a self-supervised learning (SSL) encoder in a self-supervised manner, providing a target domain enhanced signal to the SSL encoder, and providing source domain clean signals to the SSL encoder.

### Claim 17 (depends on 15)

The non-transitory computer-readable recording medium of claim 15, wherein the training the customized SE model comprises training the customized SE model based on the ensemble mapping, and the training the customized SE model further comprises pseudo labeling the noisy speech data from the target domain.

### Claim 18 (depends on 15)

The non-transitory computer-readable recording medium of claim 15, wherein the training the customized SE model comprises training the customized SE model based on the self-supervised adaptation loss, and the training the customized SE model further comprises using a distance metric in an SSRA loss term.

### Claim 19 (depends on 15)

The non-transitory computer-readable recording medium of claim 15, wherein the noisy speech data is obtained from the user device in the target domain.

### Claim 20 (depends on 19)

The non-transitory computer-readable recording medium of claim 19, wherein the user device comprises at least one of a mobile phone, a refrigerator, a smart watch, glasses, or a television.