Intelligent accompanying and guarding device based on voiceprint locking

Abstract

The application provides an intelligent accompanying device based on voiceprint locking, which comprises a main body, a control device arranged on the main body, an interactive device electrically connected with the control device, the interactive device comprising a voice recognition module and an emotional accompanying module. The emotional accompanying module can recognize voice information of a corresponding user through the voice recognition module, and output personalized emotional information according to the voice information and a user label and play the personalized emotional information. The device can realize voiceprint locking, improve the privacy and security of the device, and provide targeted personalized emotional accompanying, and enhance emotional resonance and user experience.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence devices, specifically to an intelligent companion device based on voiceprint locking. Background Technology

[0002] With the development of artificial intelligence technology, intelligent companionship devices provide emotional support and interaction for groups such as elderly people living alone and children. Voice interaction, as the primary interaction method for intelligent companionship devices, directly impacts user experience due to its level of intelligence and personalization. However, existing intelligent companionship devices still suffer from the following technical shortcomings in practical applications: First, existing smart voice devices typically use a universal, fixed wake word for voice activation, meaning anyone can activate the device simply by speaking the wake word. This wake-up method lacks uniqueness and privacy, failing to effectively restrict and lock down the user's identity, and thus failing to meet the security and privacy needs of personalized companionship.

[0003] Secondly, in scenarios involving multiple users, such as families, existing companionship devices often only provide uniform responses. Once activated, the device cannot distinguish the user being communicated with, resulting in an inability to offer targeted, personalized emotional interactions based on the user's identity, age, or habits. This lack of genuine emotional resonance significantly diminishes the effectiveness of companionship.

[0004] Therefore, there is an urgent need in this field for an intelligent companion device that can achieve dedicated secure wake-up, support automatic identification of multiple users and personalized interaction. Summary of the Invention

[0005] To overcome the existing technical problems, the present invention provides an intelligent companion device based on voiceprint locking.

[0006] The present invention adopts the following technical solution.

[0007] A smart companion device based on voiceprint locking includes a main body and a control device disposed on the main body, and also includes an interactive device electrically connected to the control device. The interactive device includes a voice recognition module and an emotional companion module. The control device has an idle state and an active state. When the control device is idle, the voice recognition module can collect the user's voice information and extract the activation voiceprint feature. The control device matches the activation voiceprint feature with several wake-up voiceprint features in the preset database. If the match is successful, the corresponding user tag is marked according to the successfully matched wake-up voiceprint feature, the control device is switched to the active state and the emotional companionship module is started. When the control device is in an active state, if the voice recognition module does not recognize the user's voice information within the preset sleep time, it will switch to an idle state and turn off the emotional companionship module. The emotional support module can recognize the voice information of the corresponding user through the voice recognition module, and output and play personalized emotional information based on the voice information and user tags.

[0008] As a further improvement of the present invention, the main body is provided with an encrypted QR code and / or the control device is provided with a unique serial number.

[0009] As a further improvement of the present invention, the database contains multiple users' wake-up voiceprint features, multiple user tags that correspond one-to-one with the wake-up voiceprint features, and multiple greetings that correspond to the user tags; When the emotional support module is started, it can play corresponding greetings based on the user's tags.

[0010] As a further improvement of the present invention, the control device is electrically connected to a storage module; The database is located in the storage module, and the wake-up voiceprint features, voice information, and personalized emotional information are encrypted and stored in the storage module.

[0011] As a further improvement of the present invention, the control device is electrically connected to a communication module; When the control device is idle, if the voice recognition module fails to recognize the user's voice information within the preset warning time, it generates an active warning signal and pushes the active warning signal to the preset terminal device or email address through the communication module.

[0012] As a further improvement of the present invention, the control device is electrically connected to a vision sensing module. The visual sensing module can monitor changes in the user's posture. When the visual sensing module determines that the user has fallen by monitoring changes in the user's posture, the control device generates a fall alarm and pushes it to a preset terminal device, email address, or medical institution through the communication module.

[0013] As a further improvement of the present invention, the main body includes multiple motion joints and multiple torsos rotatably connected by the motion joints, and the torsos and / or motion joints are provided with motion drive mechanisms. The control device can acquire road environment information from the vision sensing module and control the motion drive mechanism to work based on the road environment information.

[0014] As a further improvement of the present invention, the emotional companionship module is equipped with an artificial intelligence local model that can output personalized emotional information based on voice information and user tags, or the emotional companionship module is equipped with an intelligent interaction unit that can interact with an online artificial intelligence model through a communication module to obtain personalized emotional information.

[0015] As a further improvement of the present invention, the control device is electrically connected to a vibration breathing component, which is disposed on the main body and can drive the main body to generate rhythmic vibration.

[0016] The beneficial effects of this invention are as follows: 1. Achieve voiceprint locking to enhance device privacy and security. This overcomes the shortcomings of traditional companion devices that rely solely on a universal, fixed wake-up word and are easily activated by any user. A dedicated activation mechanism is built to ensure that only users with pre-set wake-up voiceprint characteristics can activate the device, thus enhancing its privacy and security.

[0017] 2. Provide targeted, personalized emotional support. The emotional support module can combine user tags and voice information in subsequent interactions to output and play exclusive, personalized emotional messages. This solution enables the device to automatically distinguish the specific identity of the current communication partner in multi-user scenarios such as families, and to adopt appropriate forms of address and interaction strategies that conform to their identity and habits, achieving personalized emotional interaction and significantly enhancing emotional resonance and user experience. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a diagram of the main modules of the present invention; Figure 2 This is a structural diagram of the interactive device of the present invention; Figure 3 This is a diagram of the connection module of the control device of the present invention. Detailed Implementation

[0020] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent. To better illustrate this embodiment, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product.

[0021] It will be understood by those skilled in the art that certain well-known structures and their descriptions may be omitted in the accompanying drawings. The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] Reference Figures 1 to 3 As can be seen, the present invention proposes an intelligent companion device based on voiceprint locking, including a main body and a control device disposed on the main body, and also includes an interactive device electrically connected to the control device. The interactive device includes a voice recognition module and an emotional companion module. The control device has an idle state and an active state. When the control device is idle, the voice recognition module can collect the user's voice information and extract the activation voiceprint feature. The control device matches the activation voiceprint feature with several wake-up voiceprint features in the preset database. If the match is successful, the corresponding user tag is marked according to the successfully matched wake-up voiceprint feature, the control device is switched to the active state and the emotional companionship module is started. When the control device is in an active state, if the voice recognition module does not recognize the user's voice information within the preset sleep time, it will switch to an idle state and turn off the emotional companionship module. The emotional support module can recognize the voice information of the corresponding user through the voice recognition module, and output and play personalized emotional information based on the voice information and user tags.

[0023] Voiceprint locking enhances the privacy and security of the device. It overcomes the shortcomings of traditional companion devices that rely solely on a universal, fixed wake-up word and are easily activated by any unauthorized person. A dedicated activation mechanism is built to ensure that only users with pre-set wake-up voiceprint characteristics can activate the device, thus improving privacy and security.

[0024] Furthermore, it can provide targeted, personalized emotional support. The emotional support module can combine user tags and voice information in subsequent interactions to output and play exclusive, personalized emotional messages. This solution enables the device to automatically distinguish the specific identity of the current communication partner in multi-user scenarios such as families, and to adopt appropriate forms of address and interaction strategies that conform to their identity and habits, achieving personalized emotional interaction and significantly enhancing emotional resonance and user experience.

[0025] As a further improvement of the present invention, the main body is provided with an encrypted QR code and / or the control device is provided with a unique serial number.

[0026] This invention preferably sets both an encrypted QR code and a unique serial number, assigning each device a unique physical and electronic identifier. This establishes a unique "one device, one number" identity for each device, combining hardware and software, thus constructing a reliable anti-counterfeiting and traceability system. This enables precise tracking and identity verification of the device throughout its entire lifecycle—from production and manufacturing to logistics, distribution, and retail—effectively preventing the circulation of counterfeit and substandard products at the source.

[0027] Furthermore, in subsequent after-sales processes, the unique identification of "one device, one number" avoids after-sales ownership disputes caused by the inability to identify or confuse the device's identity, thus ensuring permanent after-sales service.

[0028] As a further improvement of the present invention, the database contains multiple users' wake-up voiceprint features, multiple user tags that correspond one-to-one with the wake-up voiceprint features, and multiple greetings that correspond to the user tags; When the emotional support module is started, it can play corresponding greetings based on the user's tags.

[0029] When different members wake up the device, the emotional support module can accurately identify the user's specific identity through user tags the moment it starts up, and immediately play the corresponding greeting. For example, if the waker is a grandfather, it will immediately call out "Grandpa"; if the waker is a child, it will immediately call out their special title or name. This changes the limitation of traditional devices that can only output a uniform fixed response after being woken up.

[0030] As a further improvement of the present invention, the control device is electrically connected to a storage module; The database is located in the storage module, and the wake-up voiceprint features, voice information, and personalized emotional information are encrypted and stored in the storage module.

[0031] This invention stores wake-up voiceprint features, voice information, and personalized emotional information in a local storage module using a unified encryption algorithm. This solution ensures that even if the device is subjected to unauthorized physical disassembly, external storage access, or electronic intrusion, the plaintext wake-up voiceprint features and conversation records cannot be obtained or decrypted from the storage module, thus protecting the user's personal privacy and security.

[0032] As a further improvement of the present invention, the control device is electrically connected to a communication module; When the control device is idle, if the voice recognition module fails to recognize the user's voice information within the preset warning time, it generates an active warning signal and pushes the active warning signal to the preset terminal device or email address through the communication module.

[0033] Traditional companionship devices can only passively wait for user activation to respond. If the user is unable to speak due to emergencies such as sudden illness, disability, accidental falls, or physical discomfort, the traditional device becomes completely useless. This invention introduces a timed monitoring mechanism during idle periods, enabling the device to intelligently identify potential safety risks or abnormal interpersonal communication disruptions. This transforms the device from passively waiting to be activated to actively monitoring for risks. It establishes an active safety monitoring mechanism for abnormal unattended interaction states, truly realizing the companionship function.

[0034] As a further improvement of the present invention, the control device is electrically connected to a vision sensing module. The visual sensing module can monitor changes in the user's posture. When the visual sensing module determines that the user has fallen by monitoring changes in the user's posture, the control device generates a fall alarm and pushes it to a preset terminal device, email address, or medical institution through the communication module.

[0035] Traditional fall monitoring devices typically require users to wear hardware, manually press an emergency call button after a fall, or actively call for assistance. However, elderly people living alone often lose the ability to manually call for help due to severe impact, unconsciousness, or intense pain during a fall. This invention introduces a visual sensing module, utilizing computer vision technology to continuously and non-contactly monitor and intelligently identify changes in the user's posture in real time. Without requiring additional devices or manual operation, it automatically identifies and triggers alarms the instant a fall occurs, significantly improving the reliability and timeliness of safety monitoring. Specifically, changes in user posture can be analyzed using a directed bounding box (OBB), with the long axis representing the torso's orientation and aspect ratio reflecting the posture. The visual sensing module can use an RGB color camera to generate the OBB.

[0036] As a further improvement of the present invention, the main body includes multiple motion joints and multiple torsos rotatably connected by the motion joints, and the torsos and / or motion joints are provided with motion drive mechanisms. The control device can acquire road environment information based on the visual sensing module and control the motion drive mechanism accordingly. Similarly, road environment information can be generated through lightweight semantic segmentation based on an RGB color camera. To improve accuracy, a depth camera can be used to generate road environment information.

[0037] As a further improvement of the present invention, the emotional companionship module is equipped with an artificial intelligence local model that can output personalized emotional information based on voice information and user tags, or the emotional companionship module is equipped with an intelligent interaction unit that can interact with an online artificial intelligence model through a communication module to obtain personalized emotional information.

[0038] The local AI model can be a large model such as a package or a question. This invention preferably uses a local AI model to ensure the stable operation of the emotional companionship function even in environments without or with a weak network connection. Furthermore, it fundamentally prevents the risk of data leakage caused by uploading sensitive interactive data to the cloud. Online AI model interaction can be achieved through the model's API interface.

[0039] As a further improvement of the present invention, the control device is electrically connected to a vibration breathing component, which is disposed on the main body and can drive the main body to generate rhythmic vibration.

[0040] This invention uses a vibrating breathing component to generate rhythmic vibrations in the main body, simulating the physiological rhythmic micro-vibrations of a real living organism during its breathing and heartbeat at the hardware level. When users touch or hold the device, they can directly perceive the chest rise and fall and breathing rhythm of a real living being, extending pure language interaction to a dual simulation of "hearing and touch," enhancing the sense of life and realism of the intelligent companion device.

[0041] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A smart companion device based on voiceprint locking, comprising a main body and a control device disposed on the main body, characterized in that, It also includes an interactive device electrically connected to the control device, the interactive device including a voice recognition module and an emotional companionship module, and the control device having an idle state and an active state; When the control device is in an idle state, the voice recognition module can collect the user's voice information and extract the activation voiceprint feature. The control device matches the activation voiceprint feature with several wake-up voiceprint features in a preset database. If the match is successful, the corresponding user tag is marked according to the successfully matched wake-up voiceprint feature, the control device is switched to the active state and the emotional companionship module is started. When the control device is in an active state, if the voice recognition module does not recognize the user's voice information within a preset sleep time, it will switch to an idle state and turn off the emotional companionship module. The emotional support module can recognize the voice information of the corresponding user through the voice recognition module, and output and play personalized emotional information based on the voice information and user tags.

2. The intelligent companion device based on voiceprint locking according to claim 1, characterized in that, The main body is equipped with an encrypted QR code and / or the control device is equipped with a unique serial number.

3. The intelligent companion device based on voiceprint locking according to claim 1, characterized in that, The database contains wake-up voiceprint features of multiple users, multiple user tags that correspond one-to-one with the wake-up voiceprint features, and multiple greetings that correspond to the user tags; When the emotional support module is activated, it can play a corresponding greeting based on the user's tag.

4. The intelligent companion device based on voiceprint locking according to claim 1, characterized in that, The control device is electrically connected to a storage module; The database is located in the storage module, and the wake-up voiceprint features, voice information, and personalized emotional information are encrypted and stored in the storage module.

5. The intelligent companion device based on voiceprint locking according to claim 1, characterized in that, The control device is electrically connected to a communication module; When the control device is idle, if the voice recognition module fails to recognize the user's voice information within a preset warning period, it generates an active warning signal and pushes the active warning signal to a preset terminal device or email address through the communication module.

6. The intelligent companion device based on voiceprint locking according to claim 5, characterized in that, The control device is electrically connected to a vision sensing module; The visual sensing module can monitor changes in the user's posture. When the visual sensing module determines that the user has fallen by monitoring changes in the user's posture, the control device generates a fall alarm and pushes it to a preset terminal device, email address, or medical institution through the communication module.

7. The intelligent companion device based on voiceprint locking according to claim 6, characterized in that, The main body includes multiple motion joints and multiple torsos rotatably connected by the motion joints, and the torsos and / or motion joints are provided with motion drive mechanisms; The control device can acquire road environment information based on the visual sensing module and control the motion drive mechanism to work based on the road environment information.

8. The intelligent companion device based on voiceprint locking according to claim 5, characterized in that, The emotional support module is equipped with an artificial intelligence local model that can output personalized emotional information based on the voice information and user tags, or the emotional support module is equipped with an intelligent interaction unit that can connect and interact with an online artificial intelligence model through the communication module to obtain personalized emotional information.

9. The intelligent companion device based on voiceprint locking according to claim 1, characterized in that, The control device is electrically connected to a vibration breathing component, which is mounted on the main body and can drive the main body to produce rhythmic vibrations.

Images