Multi-mode interaction recognizer

Abstract

The utility model relates to the technical field of recognizers, and discloses a multi-mode interaction recognizer which comprises a shell, a voice recognition module, a visual recognition module and an intelligent management module are arranged in the shell, and the voice recognition module and the visual recognition module are both electrically connected with the intelligent management module; wherein the voice recognition module is used for collecting and recognizing voice interaction content of a user, converting the voice interaction content into recognizable signals and sending the recognizable signals to the intelligent management module; the visual recognition module is used for collecting and recognizing posture information of a user, converting the posture information into recognizable signals and sending the recognizable signals to the intelligent management module; and the intelligent management module is used for being connected with external execution equipment, outputting a control instruction based on the received signal content, and controlling the execution equipment connected with the intelligent management module to execute corresponding operation. According to the utility model, the interaction modes of the voice recognition module and the visual recognition module can be integrated, and multi-mode interaction fusion is realized. And meanwhile, only the shell needs to be mounted during mounting, so that the mounting process is greatly simplified.

Description

Technical Field

[0001] This utility model relates to the field of recognition technology, and in particular to a multimodal interaction recognition device. Background Technology

[0002] In current exhibition hall design and construction, exhibition technology has undergone tremendous changes. Traditional exhibition halls tend to focus on popularizing science through written displays and videos, but most people find it difficult to develop an interest in complex and dry knowledge. The addition of high-tech interactive technologies such as voice recognition, artificial intelligence, and multimedia vision has made exhibition formats more diverse. As a result, sensor devices based on these technologies are becoming increasingly widespread and diverse. By combining with exhibition items to create interactive exhibits, interactive exhibits can be achieved. For example, the application of voice recognition technology allows visitors to control multimedia equipment in the exhibition hall through voice commands, such as playing audio and video, and switching display content. Another example is the application of visual recognition technology, which, through high-definition cameras and computer vision algorithms, can count the flow of people in the exhibition hall in real time and analyze visitor movement and behavior patterns. The combined application of these technologies not only improves the interactive experience for visitors, making the visit more convenient and natural, but also enables exhibition halls to better meet the needs of visitors, greatly enhancing the interactivity and overall visitor experience of the exhibition.

[0003] However, despite the widespread application of speech recognition, artificial intelligence, and multimedia vision technologies, existing technologies still have some limitations. First, the functional modules of existing sensor carriers are relatively singular and independent, lacking a unified linkage effect, making it difficult for various functions to work collaboratively and limiting the level of intelligence in exhibition hall interaction. Second, the independent carrier format leads to a complex installation process, especially in the upgrading and renovation of old exhibition halls, making integrated installation difficult. This installation method not only affects the overall aesthetics of the exhibition hall but also increases construction difficulty and cost. Utility Model Content

[0004] To address the aforementioned technical problems, this utility model provides a multimodal interaction recognizer, which improves the integration of the recognizer, thereby enhancing the intelligence level of exhibition hall interaction and simplifying installation.

[0005] The technical solution adopted by this utility model to solve its technical problem is:

[0006] A multimodal interaction recognizer includes a housing, within which a voice recognition module, a visual recognition module, and an intelligent management module are disposed, wherein the voice recognition module and the visual recognition module are electrically connected to the intelligent management module;

[0007] The voice recognition module is used to collect and recognize the user's voice interaction content, and convert it into a recognizable signal to be sent to the intelligent management module.

[0008] The visual recognition module is used to collect and recognize the user's body posture information, and convert it into a recognizable signal to be sent to the intelligent management module;

[0009] The intelligent management module is used to connect with external execution devices, and based on the received signal content, outputs control commands to control the connected execution devices to perform corresponding operations.

[0010] Preferably, the outer casing includes a main casing and a cover; the main casing has a receiving cavity, in which the voice recognition module, the visual recognition module and the intelligent management module are all located, and the cover is placed on the main casing and closes the receiving cavity.

[0011] Preferably, the outer wall of the main housing is provided with an identification hole, and the visual recognition module includes a visual recognition module body and a lens connected to the visual recognition module body, with the lens being positioned corresponding to the identification hole.

[0012] Preferably, the accommodating cavity is provided with a mounting base, and the mounting base is provided with a mounting hole communicating with the identification hole, and the lens is inserted into the mounting hole.

[0013] Preferably, the identification hole is provided with a diaphragm for protecting the lens.

[0014] Preferably, the outer wall of the main housing is provided with a sound-permeable hole, and the voice recognition module is set corresponding to the sound-permeable hole.

[0015] Preferably, the voice recognition module includes a voice recognition module body, and the voice recognition module body is provided with multiple microphones, and the microphones are arranged one-to-one with the sound transmission holes.

[0016] Preferably, the cover is provided with an interface and a plurality of first magnets, and the interface is electrically connected to the intelligent management module.

[0017] Preferably, the main housing includes a panel opposite to the cover, a top plate and a bottom plate are respectively provided on the upper and lower sides of the panel, a side plate is connected between the top plate and the bottom plate, and a threaded hole and a second magnet are provided on the bottom plate.

[0018] Preferably, the visual recognition module is located in the middle of the outer casing, and there are two voice recognition modules, respectively located on both sides of the visual recognition module.

[0019] This utility model discloses a multimodal interaction recognizer, which, compared with the prior art, has the following advantages: By setting a voice recognition module, a visual recognition module, and an intelligent management module within a housing, and electrically connecting both the voice recognition module and the visual recognition module to the intelligent management module, the interaction methods of the voice recognition module and the visual recognition module are integrated into one unit, achieving multimodal interaction fusion. Furthermore, since all components are integrated within the housing, installation only requires installing the housing, greatly simplifying the installation process. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of the multimodal interaction recognizer of this utility model.

[0021] Figure 2 for Figure 1 A cross-sectional schematic diagram.

[0022] Figure 3 This is a rear view of the multimodal interaction recognizer of this utility model.

[0023] Figure 4 This is an exploded view of the multimodal interaction recognizer of this utility model.

[0024] The components are: 1-outer shell, 11-panel, 12-top plate, 13-bottom plate, 14-side plate, 15-cover, 16-recognition hole, 17-mounting base, 18-sound hole, 19-threaded hole, 2-voice recognition module, 21-voice recognition module body, 22-microphone, 3-visual recognition module, 31-visual recognition module body, 32-lens, 4-intelligent management module, 5-diaphragm, 6-interface, 7-first magnet, 8-second magnet. Detailed Implementation

[0025] The specific embodiments of this utility model will be further described in detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this utility model, but are not intended to limit the scope of this utility model.

[0026] like Figure 1-4 As shown, a preferred embodiment of the present invention provides a multimodal interactive recognition device, including a housing 1. The housing 1 contains a voice recognition module 2, a visual recognition module 3, and an intelligent management module 4. The voice recognition module 2 and the visual recognition module 3 are both electrically connected to the intelligent management module 4.

[0027] The voice recognition module 2 is used to collect and recognize the user's voice interaction content, and convert it into a recognizable signal to be sent to the intelligent management module, such as a microphone or voice recognizer.

[0028] The visual recognition module 3 is used to collect and recognize the user's body posture information, and convert it into a recognizable signal to be sent to the intelligent management module, such as a camera;

[0029] The intelligent management module 4 is used to connect with external execution devices (such as displays, speakers, etc.), and outputs control commands based on the received signal content to control the connected execution devices to perform corresponding operations, such as controllers.

[0030] Preferably, the visual recognition module 3 is located in the middle of the outer shell 1, there are two voice recognition modules 2, which are respectively located on both sides of the visual recognition module 3, and the intelligent management module 4 is located at the rear of the visual recognition module 3.

[0031] Based on the aforementioned technical features, the multimodal interaction recognizer integrates the interaction methods of the voice recognition module 2 and the visual recognition module 3 into a single unit by setting up a voice recognition module 2, a visual recognition module 3, and an intelligent management module 4 within the housing 1, and electrically connecting both the voice recognition module 2 and the visual recognition module 3 to the intelligent management module 4. This achieves multimodal interaction fusion. Furthermore, since all components are integrated within the housing 1, installation only requires installing the housing 1, greatly simplifying the installation process.

[0032] In this embodiment, the outer casing 1 includes a main casing and a cover 15. The main casing has a receiving cavity, in which the voice recognition module 2, the visual recognition module 3, and the intelligent management module 4 are all located. The cover 15 covers the main casing and seals the receiving cavity. The main casing includes a panel 11 opposite to the cover 15. The upper and lower sides of the panel 11 are respectively provided with a top plate 12 and a bottom plate 13. A side plate 14 connects the top plate 12 and the bottom plate 13. The bottom plate 13 is provided with a threaded hole 19 and a second magnet 8. The threaded hole 19 is preferably an M6 threaded hole, which can be externally connected to a fixing device or coupling, and can be installed and used on non-metallic equipment. The second magnet 8 can be magnetically installed on the surface of metal equipment in the exhibition hall. Thus, the bottom plate 13 can be easily connected to both metal and non-metallic materials.

[0033] Meanwhile, in practical use, the multimodal interaction recognizer can also be installed through the cover 15. Therefore, a first magnet 7 can be provided on the cover 15, allowing it to be installed on the surface of metal equipment by magnetic attraction in exhibition hall applications. Additionally, the cover 15 is provided with an interface 6, which is electrically connected to the intelligent management module 4. When connection to external execution devices is required, the execution device can be directly connected to the interface 6, making the connection simple and convenient.

[0034] In this embodiment, the panel 11 is provided with an identification hole 16. The visual recognition module 3 includes a visual recognition module body 31 and a lens 32 connected to the visual recognition module body 31. The lens 32 is disposed corresponding to the identification hole 16. Specifically, during installation, a mounting base 17 can be provided in the accommodating cavity. The mounting base 17 is provided with a mounting hole communicating with the identification hole 16. The lens 32 is inserted into the mounting hole, and the visual recognition module body 31 is fixed to the mounting base 17 by screws. In addition, a diaphragm 5 is provided in the identification hole 16 to protect the lens 32.

[0035] In this embodiment, the panel 11 is further provided with a sound-transmitting hole 18, and the voice recognition module 2 is configured corresponding to the sound-transmitting hole 18. Specifically, the voice recognition module 2 includes a voice recognition module body 21, and the voice recognition module body 21 is provided with a plurality of microphones 22, each microphone 22 corresponding to a sound-transmitting hole 18. A connecting post is provided inside the accommodating cavity, and the voice recognition module body 21 is connected to the connecting post by screws to fix the voice recognition module 2.

[0036] In summary, the multimodal interaction recognizer of this application has the following advantages:

[0037] Multimodal interaction fusion: Integrating voice recognition and visual recognition interaction methods into one, achieving multimodal interaction fusion. For example, in an exhibition hall, visitors can use voice commands to query exhibit information, while the device's visual recognition function can automatically identify the visitor's gaze focus and accurately push detailed introductions and interactive content of relevant exhibits, providing visitors with a richer and more convenient interactive experience.

[0038] Integrated Design: The equipment adopts an integrated design concept, combining core components such as sensors and processors into a compact device. This design greatly simplifies the installation process, eliminating the need for multiple independent sensor carriers and complex connection lines as required by traditional technologies, thus reducing installation difficulty and time costs.

[0039] Flexible and adaptable installation: The integrated interactive recognition device is highly adaptable and can be easily installed in various exhibits and environments within the exhibition hall. For upgrades and renovations of old exhibition halls, the device can be quickly integrated with existing display facilities without requiring large-scale renovations, saving upgrade costs while maintaining the overall aesthetics and harmony of the exhibition hall.

[0040] Enhance the interactive capabilities of the exhibition hall: Strengthen the interaction between visitors and exhibits, and interact with exhibits and display content through various means such as voice and visuals, making the visit more vivid and interesting, and improving visitor participation and satisfaction.

[0041] The above description is only a preferred embodiment of the present utility model. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of the present utility model, and these improvements and substitutions should also be considered within the protection scope of the present utility model.

Claims

1. A multi-modal interaction recognizer, characterized by: The shell is internally provided with a voice recognition module, a visual recognition module and an intelligent management module, and the voice recognition module and the visual recognition module are electrically connected with the intelligent management module. The voice recognition module is used for collecting and recognizing voice interaction content of a user and converting the voice interaction content into identifiable signals and sending the signals to the intelligent management module. The visual recognition module is used for collecting and recognizing body information of a user and converting the body information into identifiable signals and sending the signals to the intelligent management module. The intelligent management module is used for connecting with an external execution device, outputting a control instruction based on received signal content, and controlling the execution device to perform a corresponding operation.

2. The multi-modal interaction recognizer of claim 1, wherein: The shell comprises a main shell and a cover, the main shell is provided with a containing cavity, the voice recognition module, the visual recognition module and the intelligent management module are arranged in the containing cavity, and the cover is combined on the main shell and seals the containing cavity.

3. The multi-modal interaction recognizer of claim 2, wherein: An outer wall of the main shell is provided with an identification hole, the visual recognition module comprises a visual recognition module body and a lens connected with the visual recognition module body, and the lens is arranged corresponding to the identification hole.

4. The multi-modal interaction recognizer of claim 3, wherein: An installation seat is arranged in the containing cavity, the installation seat is provided with an installation hole in communication with the identification hole, and the lens is inserted into the installation hole.

5. The multi-modal interaction recognizer of claim 3, wherein: A diaphragm for protecting the lens is arranged in the identification hole.

6. The multi-modal interaction recognizer of claim 2, wherein: An outer wall of the main shell is provided with a sound transmission hole, and the voice recognition module is arranged corresponding to the sound transmission hole.

7. The multi-modal interaction recognizer of claim 6, wherein: The voice recognition module comprises a voice recognition module body, a plurality of microphones are arranged on the voice recognition module body, and the microphones are arranged corresponding to the sound transmission hole.

8. The multi-modal interaction recognizer of claim 2, wherein: An interface and a plurality of first magnets are arranged on the cover, and the interface is electrically connected with the intelligent management module.

9. The multi-modal interaction recognizer of claim 2, wherein: The main shell comprises a panel opposite to the cover, a top plate and a bottom plate are arranged on the upper side and the lower side of the panel respectively, a side plate is connected between the top plate and the bottom plate, a threaded hole and a second magnet are arranged on the bottom plate.

10. The multi-modal interaction recognizer of any one of claims 1-9, wherein: The visual recognition module is arranged in the middle of the shell, and the voice recognition module is two and arranged on both sides of the visual recognition module.

Images