# US 12,118,993 B2 · Full-duplex voice dialogue method

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

- Number: US 12,118,993 B2
- Document ID: US12118993B2
- Status: Granted
- Kind: B2
- Application number: 17/639,624
- Filed: 2019-11-25
- Published: 2022-09-15
- Granted: 2024-10-15
- Earliest claimed: 2019-09-04
- Assignees: Ai Speech Co., Ltd.
- Applicants: AI Speech Co., Ltd.
- Inventors: Jiankai Deng; Jinrui Gan
- Web page: https://patentgrep.com/patent/US12118993B2

## CPC

- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G06F 3/167: {Audio in a user interface, e.g. using voice commands for navigating, audio feedback}
- G10L 15/30: Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
- G10L 25/87: Detection of discrete points within a voice signal
- G10L 2015/221: {Announcement of recognition results}
- 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

Disclosed is a full-duplex voice dialogue method applied to a voice dialogue terminal and including recording and uploading by an awakened voice dialogue terminal audio to a cloud server for determining a reply content and a first duration of the audio analyzed for determining the reply content; receiving by the voice dialogue terminal the reply content and the first duration sent by the cloud server; determining whether the first duration is equal to a duration from the moment awakening the voice dialogue terminal to the current moment of uploading the audio; and presenting the reply content to a user if consistent. Both the reply content determined by the cloud server and the duration of the audio is acquired, and the reply content is presented to the user only when the first duration and the second duration are determined as consistent, thereby ensuring proper reply content.

## Claims

### Claim 1

A full-duplex voice dialogue method applied to a voice dialogue terminal, comprising:
recording and uploading, by the voice dialogue terminal when being awakened, audio to a cloud server such that the cloud server determines a reply content and a first duration of the audio being analyzed for determining the reply content;
receiving, by the voice dialogue terminal, the reply content and the first duration sent by the cloud server;
determining, by the voice dialogue terminal, whether the first duration is equal to a second duration, the second duration being a duration from a moment when the voice dialogue terminal is awakened to a current moment when the audio is uploaded; and
when the first duration is equal to the second duration, presenting the reply content to a user by the voice dialogue terminal.

### Claim 2 (depends on 1)

The method according to claim 1, further comprising:
when the first duration is less than the second duration, determining whether a voice endpoint is contained in a redundant recording clip of the recorded audio corresponding to the second duration compared to the recorded audio corresponding to the first duration; and
if it is determined not contained, presenting the reply content to the user; and
if it is determined contained, discarding the reply content, and waiting for the cloud server to send a new reply content and a new first duration.

### Claim 3 (depends on 2)

The method according to claim 2, wherein after the audio is recorded, the method further comprises performing voice endpoint detection on the recorded audio, and storing voice endpoint information as detected, the voice endpoint information comprising a voice start time point and a voice end time point; and
said determining whether the voice endpoint is contained in the redundant recording clip of the recorded audio corresponding to the second duration compared to the recorded audio corresponding to the first duration comprises:
querying the voice endpoint information to determine whether there is a voice start time point or a voice end time point belonging to the redundant recording clip.

### Claim 4

A full-duplex voice dialogue method applied to a cloud server, comprising:
receiving recorded audio uploaded by a voice dialogue terminal;
analyzing the received recorded audio in real time to determine a reply content corresponding to a voice content contained in the received recorded audio; and
sending to the voice dialogue terminal the reply content and a first duration of the recorded audio being analyzed for determining the reply content such that the voice dialogue terminal determines whether to present the reply content to a user based on a consistency between the first duration and a second duration, the second duration being a duration from a moment when the voice dialogue terminal is awakened to a current moment when the audio is uploaded.

### Claim 5 (depends on 4)

The method according to claim 4, further comprising:
when the first duration is less than the second duration and a voice endpoint is contained in a redundant recording clip of the received recorded audio corresponding to the second duration compared to the received recorded audio corresponding to the first duration, continuing to analyze the redundant recording clip and determine a new reply content and a new first duration; and
sending the new reply content and the new first duration to the voice dialogue terminal.

### Claim 6

A voice dialogue terminal, comprising at least one processor and a memory communicatively connected with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions, when being executed by the at least one processor, cause the at least one processor to perform the steps of:
recording and uploading, by the voice dialogue terminal when being awakened, audio to a cloud server such that the cloud server determines a reply content and a first duration of the audio being analyzed for determining the reply content;
receiving, by the voice dialogue terminal, the reply content and the first duration sent by the cloud server;
determining, by the voice dialogue terminal, whether the first duration is equal to a second duration, the second duration being a duration from a moment when the voice dialogue terminal is awakened to a current moment when the audio is uploaded; and
if the first duration is equal to the second duration, presenting the reply content to a user by the voice dialogue terminal.

### Claim 7 (depends on 6)

The voice dialogue terminal according to claim 6, wherein the at least one processor is further configured to:
when the first duration is less than the second duration, determine whether a voice endpoint is contained in a redundant recording clip of the recorded audio corresponding to the second duration compared to the recorded audio corresponding to the first duration; and
if it is determined not contained, present the reply content to the user; and
if it is determined contained, discard the reply content, and wait for the cloud server to send a new reply content and a new first duration.

### Claim 8 (depends on 7)

The voice dialogue terminal according to claim 7, wherein the at least one processor is further configured to:
after the audio is recorded, perform voice endpoint detection on the recorded audio, and store voice endpoint information as detected, the voice endpoint information comprising a voice start time point and a voice end time point; and
said determining whether the voice endpoint is contained in the redundant recording clip of the recorded audio corresponding to the second duration compared to the recorded audio corresponding to the first duration comprises:
querying the voice endpoint information to determine whether there is a voice start time point or a voice end time point belonging to the redundant recording clip.

### Claim 9 (depends on 4)

A cloud server, comprising at least one processor and a memory communicatively connected with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions, when being executed by the at least one processor, cause the at least one processor to perform the steps of the method of claim 4.

### Claim 10 (depends on 9)

The cloud server according to claim 9, wherein the at least one processor is further configured to:
when the first duration is less than the second duration and a voice endpoint is contained in a redundant recording clip of the recorded audio corresponding to the second duration compared to the recorded audio corresponding to the first duration, continue to analyze the redundant recording clip and determine a new reply content and a new first duration; and
send the new reply content and the new first duration to the voice dialogue terminal.