# US 2021/0082402 A1 · System and method for accent classification

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

- Number: US 2021/0082402 A1
- Document ID: US20210082402A1
- Status: Published application
- Kind: A1
- Application number: 16/570,122
- Filed: 2019-09-13
- Published: 2021-03-18
- Earliest claimed: 2019-09-13
- Assignees: Cerence Operating Company
- Applicants: Cerence Operating Company
- Inventors: Yang SUN; Junho PARK; Goujin WEI; Daniel WILLETT
- Web page: https://patentgrep.com/patent/US20210082402A1

## CPC

- G10L 15/07: to the speaker
- G10L 15/18: using natural language modelling
- G06F 40/30: Semantic analysis
- G06F 40/56: Natural language generation
- G10L 13/10: Prosody rules derived from text; Stress or intonation
- G10L 15/16: using artificial neural networks
- G10L 15/1807: {using prosody or stress}
- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- 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: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system and/or method receives speech input including an accent. The accent is classified with an accent classifier to yield an accent classification. Automatic speech recognition is performed based on the speech input and the accent classification to yield an automatic speech recognition output. Natural language understanding is performed on the speech recognition output and the accent classification determining an intent of the speech recognition output. Natural language generation generates an output based on the speech recognition output and the intent and the accent classification. An output is rendered using text to speech based on the natural language generation and the accent classification.

## Claims

### Claim 1

A computer-implemented method comprising:
receiving speech input, the speech including an accent;
classifying the accent with an accent classifier to yield an accent classification;
performing automatic speech recognition based on the speech input and the accent classification to yield an automatic speech recognition output;
performing natural language understanding on the speech recognition output determining an intent of the speech recognition output;
using natural language generation to generate an output based on the speech recognition output and the intent; and
rendering an output using text to speech based on the natural language generation.

### Claim 2 (depends on 1)

The computer-implemented method according to claim 1, wherein the performing natural language understanding on the speech recognition output is further based on the accent classification.

### Claim 3 (depends on 1)

The computer-implemented method according to claim 1, wherein the determining an intent is further based on the accent classification.

### Claim 4 (depends on 1)

The computer-implemented method according to claim 1, wherein using natural language generation is further based on the accent classification.

### Claim 5 (depends on 1)

The computer-implemented method according to claim 1, wherein the rendering an output is further based on the accent classification.

### Claim 6 (depends on 1)

The computer-implemented method according to claim 1, wherein the performing natural language understanding on the speech recognition output, the determining an intent, the using natural language generation, and the rendering an output are further based on the accent classification.

### Claim 7

A computer program product residing on a non-transitory computer readable storage medium having a plurality of instructions stored thereon which, when executed across one or more processors, causes at least a portion of the one or more processors to perform operations comprising:
receiving speech input, the speech including an accent;
classifying the accent with an accent classifier to yield an accent classification;
performing automatic speech recognition based on the speech input and the accent classification to yield an automatic speech recognition output;
performing natural language understanding on the speech recognition output determining an intent of the speech recognition output;
using natural language generation to generate an output based on the speech recognition output and the intent; and
rendering an output using text to speech based on the natural language generation.

### Claim 8 (depends on 7)

The computer program product according to claim 7, wherein the performing natural language understanding on the speech recognition output is further based on the accent classification.

### Claim 9 (depends on 7)

The computer program product according to claim 7, wherein the determining an intent is further based on the accent classification.

### Claim 10 (depends on 7)

The computer program product according to claim 7, wherein using natural language generation is further based on the accent classification.

### Claim 11 (depends on 7)

The computer program product according to claim 7, wherein the rendering an output is further based on the accent classification.

### Claim 12 (depends on 7)

The computer program product according to claim 7, wherein the performing natural language understanding on the speech recognition output, the determining an intent, the using natural language generation, and the rendering an output are further based on the accent classification.

### Claim 13

A computing system including one or more processors and one or more memories configured to perform operations comprising:
receiving speech input, the speech including an accent;
classifying the accent with an accent classifier to yield an accent classification; performing automatic speech recognition based on the speech input and the accent classification to yield an automatic speech recognition output;
performing natural language understanding on the speech recognition output determining an intent of the speech recognition output;
using natural language generation to generate an output based on the speech recognition output and the intent; and
rendering an output using text to speech based on the natural language generation.

### Claim 14 (depends on 13)

The computing system according to claim 13, wherein the performing natural language understanding on the speech recognition output is further based on the accent classification.

### Claim 15 (depends on 13)

The computing system according to claim 13, wherein the determining an intent is further based on the accent classification.

### Claim 16 (depends on 13)

The computing system according to claim 13, wherein using natural language generation is further based on the accent classification.

### Claim 17 (depends on 13)

The computing system according to claim 13, wherein the rendering an output is further based on the accent classification.

### Claim 18 (depends on 13)

The computing system according to claim 13, wherein the performing natural language understanding on the speech recognition output, the determining an intent, the using natural language generation, and the rendering an output are further based on the accent classification.