Granted Patent B2

US 12,211,508 · App. 17/788,591 · Granted Jan 28, 2025

Inventors: Weisi Shi (Suzhou, CN); Hongbo Song (Suzhou, CN); Chengya Zhu (Suzhou, CN); Shuai Fan (Suzhou, CN)

Assignee: AI SPEECH CO., LTD.

G10L15/30G10L15/22H04L67/141

Abstract

Claims (28)

1. A server-side processing method for actively initiating a dialogue, comprising:

establishing a communication connection with a voice client, in response to a received request for connecting from the voice client;

receiving an information stream sent by the voice client through the communication connection;

performing a dialogue decision-making process according to the information stream, obtaining and outputting an adapted dialogue content to the voice client upon determining that it is an active dialogue scenario,

wherein the performing a dialogue decision-making process according to the information stream, obtaining and outputting an adapted dialogue content to the voice client upon determining that it is an active dialogue scenario comprises

configuring a trigger condition of an active dialogue scenario and storage of dialogue content associated with the trigger condition;

determining whether it is an active dialogue scenario according to the information stream and the configured trigger condition of the active dialogue scenario, obtaining and outputting dialogue content stored in association with a current trigger condition to the voice client when the active dialogue scenario is determined,

wherein the information stream comprises audio information picked up by the voice client, the triggering condition comprises that recognition content is contained and the recognition content is invalid semantics, and the determining whether it is an active dialogue scenario according to the information stream and the configured trigger condition of the active dialogue scenario comprises

recognizing the audio information to obtain a recognition result;

determining whether the recognition result contains a recognition content, performing semantic parsing on the recognition content when the recognition content is contained, and when the semantic parsing result is invalid semantics, an active dialogue scenario is determined.

2. The server-side processing method according to claim 1 , wherein the triggering condition further comprises that recognition content is not contained while a corresponding context status exists, and the determining whether it is an active dialogue scenario according to the information stream and the configured trigger condition of the active dialogue scenario further comprises

determining whether the recognition result contains the recognition content, acquiring a context status of a voice interaction scenario for determination when the recognition content is not contained, and determining that the active dialogue scenario exists when the acquired context status of the voice interaction scenario is a corresponding context status in the trigger condition.

3. The server-side processing method according to claim 2 , wherein the corresponding context status in the trigger condition comprises waiting for inquiry and breaking silence.

4. The server-side processing method according to claim 1 , wherein the communication connection is a long connection of duplex communication.

5. An electronic device, comprising at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform steps of the method of claim 1 .

6. A non-transitory computer-readable_storage medium storing a computer program, wherein the program implements steps of the method of claim 1 when executed by a processor.

7. A server for implementing an active initiation of a dialogue, which is configured with

a communication module for establishing a communication connection with a voice client, in response to a received request for connecting from the voice client;

an information receiving module for receiving an information stream sent by the voice client through the communication connection;

a dialogue decision-making module for performing a dialogue decision-making process according to the information stream, obtaining and outputting an adapted dialogue content to the voice client when an active dialogue scenario is determined; and

a configuration module for configuring a trigger condition of an active dialogue scenario and storing dialogue content associated with the trigger condition;

the dialogue decision-making module comprises

a condition determination unit for determining whether it is an active dialogue scenario according to the information stream and the configured trigger condition of the active dialogue scenario, and calling a dialogue initiation unit when the active dialogue scenario is determined; and

the dialogue initiating unit for acquiring and outputting dialogue content stored in association with a current trigger condition to the voice client,

wherein the information stream comprises audio information picked up by the voice client, and the trigger condition comprises that the audio information contains recognition content and the recognition content is invalid semantics, and the audio information does not contain recognition content and has corresponding context status comprising waiting for inquiry and breaking silence.

8. A voice interaction system capable of actively initiating a dialogue, which comprises a voice client and a voice server-side, wherein,

the voice client is configured to initiate a connection request with the voice server-side, output collected audio information to the voice server-side in real time through the established communication connection after the communication connection is established, and play dialogue content sent by the voice server-side upon receiving the same; and

the voice server-side is the server for implementing the active initiation of the dialogue according to claim 7 .

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (41)