CN-121983052-A - Intelligent voice interactive drinking water safety consultation robot system

CN121983052ACN 121983052 ACN121983052 ACN 121983052ACN-121983052-A

Abstract

The invention relates to the technical field of robots, in particular to an intelligent voice interactive drinking water safety consultation robot system which comprises a voice interaction module, a demand recognition module and a voice consultation module, wherein the voice interaction module is used for acquiring voice data and environment perception data of a user, the demand recognition module is used for recognizing drinking water safety demands of the user associated with the current situation according to the voice data and the environment perception data, and the voice consultation module is used for initiating drinking water safety voice consultation services corresponding to the drinking water safety demands of the user through the voice interaction module, so that the defect that the traditional system cannot actively recognize and provide early warning or consultation services associated with the situation in time is effectively overcome, and preventive guarantee delay or missing caused by the fact that the best opportunity for safety intervention is possibly missed.

Inventors

YANG GUANGTAO
Cai Qinyuan
CHEN HUIYAN

Assignees

深圳市水源环保建设有限公司

Dates

Publication Date: 20260505
Application Date: 20260331

Claims (8)

1. An intelligent voice interactive drinking water safety consultation robot system, comprising: The voice interaction module is used for acquiring voice data and environment perception data of a user; The demand identification module is used for identifying the drinking water safety demand of the user associated with the current situation according to the voice data and the environment perception data; and the voice consultation module is used for initiating drinking water safety voice consultation service corresponding to the drinking water safety requirement of the user through the voice interaction module.
2. The intelligent voice interactive water safety advisory robot system of claim 1, wherein the voice interactive module comprises an audio acquisition unit and a sensor array, the step of obtaining voice data and context awareness data of a user comprising: acquiring an audio stream through the audio acquisition unit, and separating the audio stream to obtain a user voice signal and an environment background signal; Performing text conversion on the user voice signal to obtain a voice content text, and performing drinking water equipment operation sound recognition on the environment background signal to obtain an environment sound event identifier; Performing a multimodal alignment and fusion process based on the voice content text and the ambient sound event identification to generate the voice data carrying the contextual information, and Acquiring a multi-source environment signal through the sensor array, and detecting a physical operation state and an environment physical parameter related to the drinking water equipment in the multi-source environment signal to obtain environment state information; and performing multi-source information verification and compensation operation according to the environment state information and the environment sound event identifier, and generating the environment perception data with the time synchronization mark.
3. The intelligent voice interactive water safety advisory robot system of claim 2, wherein said step of identifying user water safety requirements associated with a current scenario from said voice data and said context awareness data comprises: performing multi-mode fusion analysis according to the voice data and the environment perception data, and identifying drinking water related activities of the current scene to obtain drinking water activity characteristics; and acquiring the drinking water safety requirement of the user in the drinking water related activities according to the drinking water activity characteristics and the voice content text.
4. The intelligent voice interactive drinking safety advisory robot system as claimed in claim 3, wherein the step of performing a multimodal fusion analysis based on the voice data and the environmental awareness data, identifying drinking related activities currently being performed by the user, and obtaining drinking activity characteristics comprises: respectively extracting a voice time sequence characteristic from the voice data and extracting an environment time sequence characteristic from the environment sensing data; Performing time sequence alignment and feature complementary fusion on the voice time sequence features and the environment time sequence features through an attention fusion network to generate a fusion feature sequence; Identifying the type of the drinking water activity by the fusion characteristic sequence to obtain a drinking water activity type identification result; Performing scene modeling based on the drinking water activity type identification result and context information related to time, space and equipment state in the fusion feature sequence to generate a scene context; combining and structuring the drinking water activity type identification result and the context, and outputting drinking water activity information of the drinking water activity type and the context; Wherein the drinking activity feature comprises the drinking activity type and the drinking activity information.
5. The intelligent voice interactive water safety advisory robot system as claimed in claim 4, wherein said step of determining said user's water safety requirement present in said water related activity based on said water activity characteristics and said voice content text comprises: performing demand reasoning according to the drinking water activity characteristics and the voice content text, and generating a drinking water safety demand set associated with the drinking water related activity; Carrying out risk relevance verification and importance sequencing on each drinking water safety requirement in the drinking water safety requirement set through a preset drinking water safety knowledge graph to obtain a requirement list; according to the real-time physiological state and emotion tendencies of the user included in the voice content text, calculating the urgency and matching degree of each drinking water safety requirement in the requirement list under the current situation; And based on the urgency and the matching degree, carrying out weighted fusion and reordering on the requirement list to generate the drinking water safety requirement of the user.
6. The intelligent voice interactive drinking water safety counseling robot system according to claim 5, wherein the step of initiating a drinking water safety voice counseling service corresponding to the user's drinking water safety requirement through the voice interactive module comprises: matching a drinking water safety service strategy corresponding to the drinking water safety requirement of the user in a preset multi-dimensional service strategy library; Generating consultation contents based on the drinking water safety service strategy and the environment perception data; invoking a preset voice synthesis model, and generating personalized voice streams according to the consultation content, the historical interaction preference and the user acoustic characteristics in the voice content text; and broadcasting the personalized voice stream to the user through the voice interaction module so as to initiate the drinking water safety voice consultation service.
7. The intelligent voice interactive water safety advisory robot system as claimed in claim 6, wherein said step of broadcasting said personalized voice stream to said user to initiate said water safety voice advisory service comprises, after: and adjusting the detail degree and broadcasting rhythm of the consultation content corresponding to the personalized voice stream according to the user feedback voice information acquired in the broadcasting process.
8. The intelligent voice interactive drinking safety counseling robot system according to claim 7, wherein the step of adjusting the detail level of the counseling contents and the broadcasting rhythm corresponding to the personalized voice stream according to the user feedback voice information collected during the broadcasting process comprises: performing voice recognition and emotion analysis on the user feedback voice information to obtain a voice feedback text and emotion state characteristics; Acquiring feedback intention of the user based on the voice feedback text and the emotion state characteristics in a preset interaction intention classification model; After the feedback intention is matched to obtain a corresponding content detail degree adjustment strategy and a broadcasting rhythm adjustment strategy, calculating to obtain a content detail degree parameter and a broadcasting rhythm parameter by combining the historical interaction preference; And generating an adjusted personalized voice stream according to the content detail degree parameter and the broadcasting rhythm parameter.

Description

Intelligent voice interactive drinking water safety consultation robot system Technical Field The invention relates to the technical field of robots, in particular to an intelligent voice interactive drinking water safety consultation robot system. Background The drinking water safety is directly related to public health, and timely and accurate drinking water safety consultation can help people to know water quality conditions, identify potential risks and take correct countermeasures, so that drinking water safety and physical health are effectively guaranteed. At present, an intelligent voice question-answering system is generally used for providing convenient query of drinking water safety information for users, the users can actively inquire specific questions through voice, and the system gives corresponding voice answers based on a knowledge base, so that information retrieval interaction is realized. However, such systems typically rely on the user itself to recognize the problem and to actively initiate the query, with the service mode essentially being a passive response of "one-to-one", lacking the ability to actively perceive and judge the specific scenario in which the user is located and the potential drinking water safety needs that may exist that are not explicitly expressed, and when the user is unaware of the existence of the risk due to lack of relevant knowledge or is flawless of active queries due to being in a particular drinking water activity scenario, the existing systems cannot actively identify and provide timely, scenario-related early warning or advisory services, potentially missing the best opportunity to conduct safety interventions, which makes their preventive safeguarding effect much compromised. Disclosure of Invention The application provides an intelligent voice interactive drinking water safety consultation robot system, which aims to solve the technical problem that the preventive guarantee effect of the existing intelligent voice drinking water safety consultation system is defective. The intelligent voice interactive drinking safety consultation robot system provided by the application adopts the following technical scheme: an intelligent voice interactive drinking water safety consultation robot system, comprising: The voice interaction module is used for acquiring voice data and environment perception data of a user; the demand identification module is used for identifying the drinking water safety demand of the user associated with the current situation according to the voice data and the environment perception data; the voice consultation module is used for initiating drinking water safety voice consultation service corresponding to the drinking water safety requirement of the user through the voice interaction module. Further, the voice interaction module comprises an audio acquisition unit and a sensor array, and the step of acquiring voice data and environment perception data of the user comprises the following steps: acquiring an audio stream through an audio acquisition unit, and separating the audio stream to obtain a user voice signal and an environment background signal; Performing text conversion on the voice signal of the user to obtain a voice content text, and performing drinking water equipment operation sound recognition on the environmental background signal to obtain an environmental sound event identifier; based on the voice content text and the environmental sound event identifier, performing multi-mode alignment and fusion processing to generate voice data carrying context associated information, collecting multi-source environmental signals through a sensor array, and detecting physical operation states and environmental physical parameters related to drinking water equipment in the multi-source environmental signals to obtain environmental state information; And performing multi-source information verification and compensation operation according to the environment state information and the environment sound event identifier to generate environment perception data with a time synchronization mark. Further, the step of identifying the user hydration safety requirement associated with the current scenario based on the voice data and the context awareness data comprises: performing multi-mode fusion analysis according to the voice data and the environment perception data, and identifying drinking water related activities of the current scene to obtain drinking water activity characteristics; And acquiring the drinking water safety requirements of the user in the drinking water related activities according to the drinking water activity characteristics and the voice content text. Further, performing multi-mode fusion analysis according to the voice data and the environment perception data, and identifying the drinking water related activities currently performed by the user, wherein the step of obtaining the drinking water activity characteristi