# US 2026/0311716 A1 · Method and device for generating speech enhancement model, and speech enhancement method and device

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

- Number: US 2026/0311716 A1
- Document ID: US20260311716A1
- Status: Published application
- Kind: A1
- Application number: 18/725,147
- Filed: 2022-12-13
- Published: 2026-10-08
- Earliest claimed: 2021-12-28
- Assignees: Orka Health Co., Ltd.
- Applicants: Orka Health Co., Ltd.
- Inventors: Congxi LU; Linkai LI; Haoshuai ZHOU; Yufan YUAN; Hongcheng SUN
- Web page: https://patentgrep.com/patent/US20260311716A1

## CPC

- G10L 21/0208: Noise filtering
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present application discloses a method for generating a speech enhancement model, including: obtaining audio training data; obtaining a first model and a second model, where the first model is a deep neural network model, and resources occupied by the second model during operation is less than that occupied by the first model; training the first model and the second model based on the audio training data, where the training includes: obtaining first audio data having a frame length of M from the input audio; inputting the first audio data into the first model for processing to obtain a first output result; obtaining second audio data having a frame length of N from the input audio, where the second audio data is subsequent to the first audio data, and N<M; inputting the second audio data and the first output result into the second model for processing to obtain a second output result; and updating parameters of the first model and parameters of the second model based on the second output result and output audio to obtain a trained first model and a trained second model; and generating a speech enhancement model based on the trained first model and the trained second model.

## Claims

### Claim 1

A method for generating a speech enhancement model, comprising:
obtaining audio training data, wherein the audio training data comprises noisy input audio data and noiseless output audio data corresponding to the noisy input audio data;
obtaining a first model and a second model, wherein the first model is a deep neural network model, and resources occupied by the second model during operation is less than that occupied by the first model;
training the first model and the second model based on the audio training data, wherein the training comprises:
obtaining first audio data having a frame length of M from the input audio data:
inputting the first audio data into the first model for processing to obtain a first output result;
obtaining second audio data having a frame length of Nfrom the input audio data, wherein the second audio data is subsequent to the first audio data, and N<M;
inputting the second audio data and the first output result into the second model for processing to obtain a second output result; and
updating parameters of the first model and parameters of the second model based on the second output result and the output audio data to obtain the trained first model and the trained second model; and
generating a speech enhancement model based on the trained first model and the trained second model.

### Claim 2 (depends on 1)

The method of claim 1, wherein the second model is a deep neural network model, and the parameters of the second model are fewer than the parameters of the first model.

### Claim 3 (depends on 2)

The method of claim 2, wherein a ratio of a quantity of the parameters of the second model to a quantity of the parameters of the first model is less than 1/2.

### Claim 4 (depends on 2)

The method of claim 2, wherein the first model is a convolutional recurrent neural network model, and the second model is an RNNoise model.

### Claim 5 (depends on 2)

The method of claim 2, wherein training the first model and the second model based on the audio training data further comprises:
obtaining third audio data with a frame length of P previous to the second audio data from the input audio data; and
combining the third audio data with the second audio data to form fourth audio data having a frame length of P+N,
wherein inputting the second audio data and the first output result into the second model for processing to obtain the second output result comprises:
inputting the fourth audio data and the first output result into the second model for processing to obtain an intermediate output result having a frame length of P+N; and
obtaining a result having a length of N and corresponding to the second audio data from the intermediate output result to serve as the second output result.

### Claim 6 (depends on 1)

The method of claim 1, wherein the second model comprises a digital signal processing model and a simplified neural network model.

### Claim 7 (depends on 6)

The method of claim 6, wherein the second model comprises a power spectrum estimation model and a fully connected layer.

### Claim 8 (depends on 6)

The method of claim 6, wherein inputting the second audio data and the first output result into the second model for processing to obtain the second output result comprises:
inputting the second audio data into the digital signal processing model for processing to obtain an intermediate output result; and
inputting the intermediate output result and the first output result to the simplified neural network model for processing to obtain the second output result.

### Claim 9 (depends on 1)

The method of claim 1, wherein M=K×N, and K is an integer greater than 1; and
wherein, during training the first model and the second model based on the audio training data, K consecutive pieces of second audio data each having a frame length of N subsequent to the first audio data are obtained from the input audio data, the K pieces of second audio data are sequentially inputted, together with the first output result, into the second model for processing to obtain K second output results, and the parameters of the first model and the parameters of the second model are updated based on the K second output results and the output audio data to obtain the trained first model and the trained second model

### Claim 10

(canceled)

### Claim 11 (depends on 1)

The method of claim 1, wherein a step of training the first model and the second model based on the audio training data is executed iteratively for a plurality of times, wherein in an iterative execution process, updated values of the parameters of the trained first model and updated values of the parameters of the trained second model that are generated in a previous training are respectively assigned to the first model to be trained and second model to be trained for using in a subsequent training.

### Claim 12

(canceled)

### Claim 13

(canceled)

### Claim 14

A speech enhancement method, comprising:
obtaining an audio data;
obtaining a speech enhancement model, wherein the speech enhancement model comprises a first model and a second model, the first model is a deep neural network model, and resources occupied by the second model during operation is less than that occupied by the first model;
processing the audio data by using the speech enhancement model to reduce or remove a noise signal in the audio data, wherein the processing comprises:
obtaining first audio data having a frame length of M from the audio data;
inputting the first audio data into the first model for processing to obtain a first output result;
obtaining a second audio data having a frame length of N from the input audio, wherein the second audio data is subsequent to the first audio data, and N<M; and
inputting the second audio data and the first output result into the second model for processing to obtain a second output result; and
outputting the second output result as an enhanced audio data.

### Claim 15 (depends on 14)

The speech enhancement method of claim 14, wherein the second model is a deep neural network model, and parameters of the second model are fewer than parameters of the first model.

### Claim 16 (depends on 15)

The speech enhancement method of claim 15, wherein a ratio of a quantity of the parameters of the second model to a quantity of the parameters of the first model is less than ½.

### Claim 17 (depends on 15)

The speech enhancement method of claim 15, wherein the first model is a convolutional recurrent neural network model, and the second model is an RNNoise model.

### Claim 18 (depends on 15)

The speech enhancement method of claim 15, wherein processing the audio data by using the speech enhancement model to reduce or remove the noise signal in the audio data further comprises:
obtaining third audio data with a frame length of P previous to the second audio data from the input audio; and
combining the third audio data with the second audio data to form fourth audio data having a frame length of P+N,
wherein inputting the second audio data and the first output result into the second model for processing to obtain the second output result comprises:
inputting the fourth audio data and the first output result into the second model for processing to obtain an intermediate output result having a frame length of P+N; and
obtaining a result having a length of N and corresponding to the second audio data from the intermediate output result to serve as the second output result.

### Claim 19 (depends on 14)

The speech enhancement method of claim 14, wherein the second model comprises a digital signal processing model and a simplified neural network model.

### Claim 20 (depends on 19)

The speech enhancement method of claim 19, wherein the second model comprises a power spectrum estimation model and a fully connected layer.

### Claim 21 (depends on 19)

The speech enhancement method of claim 19, wherein inputting the second audio data and the first output result into the second model for processing to obtain the second output result comprises:
inputting the second audio data into the digital signal processing model for processing to obtain an intermediate output result; and
inputting the intermediate output result and the first output result to the simplified neural network model for processing to obtain the second output result.

### Claim 22 (depends on 14)

The speech enhancement method of claim 14, wherein M=K×N, and K is an integer greater than 1; and
wherein during processing the audio data by using the speech enhancement model, K consecutive pieces of second audio data each having a frame length of N subsequent to the first audio data are obtained from the audio data, and the K pieces of second audio data are sequentially inputted, together with the first output result, into the second model for processing to obtain K second output results.

### Claim 23

(canceled)

### Claim 24 (depends on 14)

A speech enhancement device, comprising:
a processor; and
a storage apparatus configured to store a computer program executable by the processor,
wherein the computer program, when executed by the processor, causes the processor to perform the speech enhancement method of claim 14.

### Claim 25

(canceled)
