Intelligent terminal system and device with dialect speech recognition

Abstract

The present invention provides an intelligent terminal system and device with dialect speech recognition, which relates to the field of speech recognition technology. The intelligent terminal system with dialect speech recognition includes: a speech collection module, a central processing module, a data storage module, a power supply module, a network module and a speech broadcast module; the speech collection module includes a speech management unit and a speech recognition unit; the speech recognition unit collects the user's real-time dialect speech through a terminal device; the intelligent terminal device is used for speech collection and text conversion of the intelligent terminal system with dialect speech recognition. The scheme ultimately realizes that the real-time dialect features are automatically matched with the preset dialect feature group through the ASR model; when it is determined to be a known dialect feature, the corresponding dialect feature group is quickly matched; when it is determined to be an unknown dialect, the unknown dialect is quickly matched with the closest core dialect, and the closest core dialect is used for recognition.

Description

Technical Field

[0001] The present invention relates to the field of speech recognition technology, and in particular to an intelligent terminal system and device with dialect speech recognition. Background Art

[0002] In order to facilitate the recognition and analysis of language, voice control technology is introduced in smart terminals, but most smart devices can only recognize standard Mandarin. Different regions face complex dialect situations. Regional differences in dialects are common, and they are often spread in spoken form. There is a lack of corresponding text and it is difficult to collect. There is little dialect corpus data available for model training. There is a certain degree of similarity between dialects, but there are usually large differences at the language level. Therefore, it is very important to develop a speech recognition system that can recognize dialects.

[0003] In the related prior art, when a common speech recognition system performs dialect recognition, the user is required to enter voice commands and corresponding operation commands; then, after obtaining the voice information, the voice commands in the voice information are extracted, and then the corresponding operation commands are issued according to the voice command matching.

[0004] When using traditional speech recognition technology, when faced with speech in an unknown dialect, the speech recognition system will recognize and analyze it according to the known dialect, resulting in differences between the analyzed language content and the actual unknown dialect, making it inconvenient to automatically and quickly match and recognize the unknown dialect.

[0005] Therefore, it is necessary to provide an intelligent terminal system with dialect speech recognition to solve the above technical problems. Summary of the invention

[0006] The present invention provides an intelligent terminal system with dialect speech recognition, which solves the problem in the related art that it is inconvenient to automatically and quickly match and recognize unknown dialects.

[0007] In order to solve the above technical problems, the present invention provides an intelligent terminal system with dialect speech recognition, comprising:

[0008] Voice collection module, central processing module, data storage module, power supply module, network module and voice broadcast module;

[0009] The speech collection module includes a speech management unit and a speech recognition unit. The speech management unit is used to construct and update the dialect feature group. The speech recognition unit collects the user's real-time dialect speech through the terminal device.

[0010] The data storage module is used for storing the dialect feature group;

[0011] The central processing module performs feature extraction, feature matching and language conversion on the real-time dialect speech. When the user uses dialect input, the central processing module quickly extracts the real-time dialect features and matches the real-time dialect features with the dialect feature group. After the matching is completed, the real-time dialect features are automatically converted into Mandarin features to provide support for the understanding and recognition of the central processing module;

[0012] The power supply module is used for powering the equipment;

[0013] The network module is used for network access of the device;

[0014] The voice broadcast module is used for the voice content played by the broadcast system.

[0015] Preferably, the voice management unit is used for the administrator to perform input management and manual update management on the dialect feature group stored in the data storage module.

[0016] Preferably, the data storage module is integrated and installed in a server management system, and the server management system provides service support of the dialect feature group for remote terminal devices through a network module.

[0017] Preferably, the speech recognition unit is used to collect the user's real-time speech in real time. The real-time speech can be Mandarin, dialect, or English.

[0018] Preferably, the central processing module includes a speech preprocessing unit, a feature extraction unit and a feature matching unit. The speech preprocessing unit is used to perform noise reduction processing on the collected real-time speech to obtain clear speech; the feature extraction unit is used to extract dialect features in the clear speech; the feature matching unit quickly matches the dialect features with a preset dialect feature group to determine which dialect the dialect features belong to, thereby providing support for language recognition and text conversion.

[0019] Preferably, the central processing module further comprises a database construction unit and a data updating unit, wherein the database construction unit collects the dialect feature group in the big data through the network module, and matches the dialect feature group in the big data with the dialect feature group in the data storage module; and determines whether an update is required;

[0020] When an update is required, the data updating unit automatically and adaptively updates the dialect feature group in the data storage module with the dialect feature group in the big data.

[0021] The present invention also provides an intelligent terminal device, which is used for voice collection and text conversion of the intelligent terminal system with dialect voice recognition.

[0022] Preferably, the invention comprises: an all-in-one machine, the all-in-one machine comprises a housing, a touch screen, a microphone, a playback hole and a camera; a rotating opening and a buffer slot are provided on the housing, the rotating opening is communicated with the buffer slot; the touch screen, the microphone and the camera are integrated and installed on the housing, and the housing is provided with a playback hole;

[0023] An adjusting mechanism, wherein the adjusting mechanism comprises a driving member, a connecting assembly and a telescopic assembly, wherein the driving member is fixedly arranged in the housing, the connecting assembly comprises a connecting shaft, a first bevel gear, a second bevel gear and a rotating shaft, the top end of the connecting shaft is fixedly connected to the shaft end of the driving member, the bottom end of the connecting shaft is fixedly connected to the first bevel gear, the second bevel gear is rotatably mounted on the housing through the rotating shaft, the second bevel gear is meshingly connected to the first bevel gear, the telescopic assembly comprises a connecting gear and a connecting tooth plate, the connecting gear is fixedly arranged on the connecting shaft, and the connecting tooth plate is meshingly connected to the connecting gear;

[0024] A support arm, one end of which is fixedly connected to the rotating shaft, the other end of which passes through the rotating opening and is supported on a table, and the rotation range of the support arm matches the size of the rotating opening;

[0025] a first sealing plate, the first sealing plate being slidably mounted in the housing, the connecting tooth plate being fixedly connected to the first sealing plate, and the sliding range of the first sealing plate matching the size of the playback hole;

[0026] A closing device, the closing device comprising a telescopic member and a second sealing plate, the telescopic member is installed in the buffer groove, one end of the second sealing plate is inserted into the buffer groove and fixedly connected to the telescopic end of the telescopic member;

[0027] Wherein, the rotating port, the buffer groove, the driving member, the connecting assembly, the supporting arm and the closing device are each provided with two groups, and are arranged in a one-to-one correspondence.

[0028] Preferably, two playback holes are provided, and the two playback holes are symmetrically arranged on both sides of the touch screen; the playback hole, the first sealing plate and the connecting shaft are arranged in a one-to-one correspondence.

[0029] Preferably, the telescopic member is a spring telescopic structure, and both ends of the telescopic member are elastically connected to the housing and the second sealing plate respectively; the second sealing plate is provided with a slope structure facing the interior of the housing.

[0030] Compared with the related art, the intelligent terminal system with dialect speech recognition provided by the present invention has the following beneficial effects:

[0031] Real-time dialect features are obtained by extracting voice features from real-time voice data; real-time dialect features are automatically matched with preset dialect feature groups through the ASR model; when it is determined to be a known dialect feature, the corresponding dialect feature group is quickly matched; when it is determined to be an unknown dialect, the unknown dialect is quickly matched with the closest core dialect, and the closest core dialect is used for recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0033] Figure 1 A system block diagram of an intelligent terminal system with dialect speech recognition provided by the present invention;

[0034] Figure 2 for Figure 1 The system block diagram of the voice acquisition module shown;

[0035] Figure 3 for Figure 1 The system block diagram of the central processing module shown;

[0036] Figure 4 A three-dimensional diagram of a smart terminal device provided by the present invention;

[0037] Figure 5 for Figure 1 A top view of the AA section is shown;

[0038] Figure 6 for Figure 5 An enlarged schematic diagram of part B is shown;

[0039] Figure 7 for Figure 5 Rear view of the CC section shown;

[0040] Figure 8 for Figure 7 Left side view of the DD section shown;

[0041] Fig. 9 The support arm storage principle diagram of the terminal device provided by the present invention, wherein: Fig. 9 (a) is the right side view of the support arm in the unfolded state. Fig. 9 (b) is the right view of the support arm during its retraction process. Fig. 9 (c) is a right side view of the support arm in the folded state;

[0042] Fig.10 The schematic diagram of closing the playback hole of the terminal device provided by the present invention, wherein: Fig.10 (a) is the rear view with the playback hole open. Fig.10 (b) is the rear view when the playback hole is closed. Fig.10 (c) is a rear view of the playback hole in a closed state.

[0043] Description of Figure Numbers:

[0044] 1. All-in-one machine; 11. Casing; 111. Rotating port; 112. Buffer slot; 12. Touch screen; 13. Sound pickup; 14. Playing hole; 15. Camera;

[0045] 2. Adjustment mechanism; 21. Driving member; 22. Connecting assembly; 221. Connecting shaft; 222. First bevel gear; 223. Second bevel gear; 224. Rotating shaft; 23. Telescopic assembly; 231. Connecting gear; 232. Connecting tooth plate;

[0046] 3. Support arm;

[0047] 4. The first sealing plate;

[0048] 5. Closing device; 51. Telescopic member; 52. Second sealing plate.

[0049] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The invention provides an intelligent terminal system with dialect speech recognition.

[0052] Please refer to Figures 1 to 3 In a first embodiment of the present invention, an intelligent terminal system with dialect speech recognition includes:

[0053] Voice collection module, central processing module, data storage module, power supply module, network module and voice broadcast module;

[0054] The speech collection module includes a speech management unit and a speech recognition unit. The speech management unit is used to construct and update the dialect feature group. The speech recognition unit collects the user's real-time dialect speech through the terminal device.

[0055] The data storage module is used for storing the dialect feature group;

[0056] The central processing module performs feature extraction, feature matching and language conversion on the real-time dialect speech. When the user uses dialect input, the central processing module quickly extracts the real-time dialect features and matches the real-time dialect features with the dialect feature group. After the matching is completed, the real-time dialect features are automatically converted into Mandarin features to provide support for the understanding and recognition of the central processing module;

[0057] The power supply module is used for powering the equipment;

[0058] The network module is used for network access of the device;

[0059] The voice broadcast module is used for the voice content played by the broadcast system.

[0060] In this embodiment, the speech management unit is used to construct an ASR model for dialect speech recognition, and the ASR model records known dialect data in the data storage module; the ASR model uses deep learning technology to automatically recognize dialect speech and convert it into text.

[0061] The ASR model is used to recognize dialect speech; the value representing the distance is obtained by weighted calculation of the pronunciation of different dialects;

[0062] When a known dialect appears, the ASR model can quickly match the known dialect with the core dialect;

[0063] When an unknown dialect appears, the ASR model can quickly determine the distance between the unknown dialect and the core dialect, and use the closest core dialect for identification.

[0064] Real-time dialect features are obtained by extracting voice features from real-time voice data; real-time dialect features are automatically matched with preset dialect feature groups through the ASR model; when it is determined to be a known dialect feature, the corresponding dialect feature group is quickly matched; when it is determined to be an unknown dialect, the unknown dialect is quickly matched with the closest core dialect, and the closest core dialect is used for recognition.

[0065] Please refer again Figure 2 The voice management unit is used for the management personnel to perform input management and manual update management on the dialect feature group stored in the data storage module.

[0066] It is convenient for maintenance personnel to perform online update management of the background data of the dialect feature group in the data storage module. After the online update, there is no need to manually input it on each terminal device separately; the terminal device only needs to update the software system to apply the latest dialect feature group.

[0067] In this embodiment, the data storage module is integrated and installed in the server management system, and the server management system provides service support of the dialect feature group for the remote terminal equipment through the network module.

[0068] The speech recognition unit is used to collect the user's real-time speech in real time. The real-time speech can be Mandarin, dialect, or English. The languages ​​of different countries can also be expanded according to needs.

[0069] Please refer again Figure 3 The central processing module includes a speech preprocessing unit, a feature extraction unit and a feature matching unit. The speech preprocessing unit is used to perform noise reduction processing on the collected real-time speech to obtain clear speech; the feature extraction unit is used to extract dialect features in the clear speech; the feature matching unit quickly matches the dialect features with the preset dialect feature group, determines which dialect the dialect features belong to, and provides support for language recognition and text conversion.

[0070] Reduce noise interference during speech input so that there is no interference with speech content during feature extraction, making dialect recognition accurate and reliable.

[0071] In this embodiment, the central processing module, the network module, the power supply module, the voice recognition unit and the voice playback module are integrated and installed in the terminal device. The central processing module on the terminal device is equipped with an independent storage device for software installation and system storage.

[0072] Please refer again Figure 3 , the central processing module also includes a database construction unit and a data updating unit, the database construction unit collects the dialect feature group in the big data through the network module, and matches the dialect feature group in the big data with the dialect feature group in the data storage module; determines whether it needs to be updated;

[0073] When an update is required, the data updating unit automatically and adaptively updates the dialect feature group in the data storage module with the dialect feature group in the big data.

[0074] It is convenient for the dialect feature group preset in the data storage module to be adaptively updated according to the update of big data, and automatic update support is provided for the data update of the terminal device.

[0075] Please refer again Figure 1The intelligent terminal system with dialect speech recognition also includes an interface module, which is used for the terminal device to connect to the peripheral device.

[0076] It is convenient to provide support for terminal devices to connect to peripherals, so as to facilitate the convenient use of the equipment.

[0077] The recognition method of the intelligent terminal system with dialect speech recognition provided by the present invention:

[0078] S1, before use, the voice management unit enters the language management mode, collects the basic language model, and constructs the language model and stores it in the database;

[0079] S2, when in use, the speech recognition unit enters the speech recognition mode, collects the user's speech, and obtains clear speech after speech preprocessing;

[0080] S3, during processing, the feature extraction unit extracts dialect features from the clear speech;

[0081] S4, during processing, the feature matching unit quickly matches the dialect feature group corresponding to the dialect feature, recognizes the user's dialect and converts it into text.

[0082] The invention also provides an intelligent terminal device.

[0083] Please refer to Figures 4 to 7 , the intelligent terminal device is used for voice collection and text conversion of the intelligent terminal system with dialect speech recognition;

[0084] The intelligent terminal device comprises:

[0085] The all-in-one machine 1 comprises a housing 11, a touch screen 12, a microphone 13, a playback hole 14 and a camera 15; the housing 11 is provided with a rotating opening 111 and a buffer slot 112, the rotating opening 111 is connected to the buffer slot 112; the touch screen 12, the microphone 13 and the camera 15 are integrally mounted on the housing 11, and the housing 11 is provided with a playback hole 14;

[0086] An adjusting mechanism 2, wherein the adjusting mechanism 2 comprises a driving member 21, a connecting assembly 22 and a telescopic assembly 23, wherein the driving member 21 is fixedly arranged in the housing 11, and the connecting assembly 22 comprises a connecting shaft 221, a first bevel gear 222, a second bevel gear 223 and a rotating shaft 224, wherein the top end of the connecting shaft 221 is fixedly connected to the shaft end of the driving member 21, and the bottom end of the connecting shaft 221 is fixedly connected to the first bevel gear 222, and the second bevel gear 223 is rotatably mounted on the housing 11 via the rotating shaft 224, and the second bevel gear 223 is meshingly connected to the first bevel gear 222, and the telescopic assembly 23 comprises a connecting gear 231 and a connecting toothed plate 232, wherein the connecting gear 231 is fixedly arranged on the connecting shaft 221, and the connecting toothed plate 232 is meshingly connected to the connecting gear 231;

[0087] A support arm 3, one end of which is fixedly connected to the rotating shaft 224, and the other end of which passes through the rotating opening 111 and is supported on a table, and the rotation range of the support arm 3 matches the size of the rotating opening 111;

[0088] A first sealing plate 4, wherein the first sealing plate 4 is slidably mounted in the housing 11, the connecting tooth plate 232 is fixedly connected to the first sealing plate 4, and the sliding range of the first sealing plate 4 matches the size of the playback hole 14;

[0089] The closing device 5 comprises a telescopic member 51 and a second sealing plate 52. The telescopic member 51 is installed in the buffer groove 112. One end of the second sealing plate 52 is inserted into the buffer groove 112 and fixedly connected to the telescopic end of the telescopic member 51.

[0090] There are two groups of the rotating opening 111 , the buffer groove 112 , the driving member 21 , the connecting assembly 22 , the supporting arm 3 and the closing device 5 , which are arranged in a one-to-one correspondence.

[0091] In this embodiment, the driving member 21 is a motor structure, which provides a power source for the rotation adjustment of the connecting shaft 221 .

[0092] Please refer again Figure 8 In this embodiment, the two groups of support arms 3 are symmetrically installed on the back of the housing 11, and the two groups of support arms 3 are synchronously driven and rotated by the driving member 21. The two driving members 21 are synchronously controlled by the controller to achieve synchronous expansion or synchronous storage of the two support arms 3.

[0093] The driving member 21 is used to conveniently drive the connecting shaft 221 to rotate and adjust. The connecting shaft 221 can drive the first bevel gear 222 to rotate while rotating and adjusting. The first bevel gear 222 drives the second bevel gear 223 to rotate when rotating. The second bevel gear 223 can synchronously drive the support arm 3 to rotate and adjust through the rotating shaft 224, so as to realize automatic deployment or automatic folding of the support arm 3. When the support arm 3 is folded, the connecting shaft 221 also drives the connecting gear 231 to rotate. When the connecting gear 231 rotates, it drives the connecting tooth plate 232 to move. When the connecting tooth plate 232 moves, it drives the first sealing plate 4 to move and close the first sealing plate 4 within the range of the playback hole 14.

[0094] After the support arm 3 is retracted, the telescopic member 51 can drive the second sealing plate 52 to be telescopically adjusted, so as to realize automatic closing of the rotating opening 111;

[0095] Finally, when the support arm 3 switches from the extended state to the retracted state, the first cover plate 4 is automatically closed. After the first cover plate 4 is closed, the second cover plate 52 can be automatically closed by the telescopic member 51. This allows the support arm 3 to be automatically retracted, the playback hole 14 to be automatically closed, and the rotating mouth 111 to be closed when the device is not needed, thereby reducing pollution to the external environment when the device is retracted.

[0096] Please combine Figure 4 and Figure 7 There are two playback holes 14 , and the two playback holes 14 are symmetrically arranged on both sides of the touch screen 12 ; the playback holes 14 , the first sealing plate 4 and the connecting shaft 221 are arranged in a one-to-one correspondence.

[0097] Two groups of symmetrical playback holes 14 are provided to increase the sound playback effect of the housing 11; while the two support arms 3 are rotated and adjusted, the two first sealing plates 4 can be controlled to be adjusted to provide support for the dual speakers of the device.

[0098] In an optional implementation of the present embodiment, the telescopic member 51 can be an electric telescopic rod, the fixed end of the electric telescopic rod is fixed on the housing 11 and is located in the buffer groove 112, and the second sealing plate 52 is fixed at the telescopic end of the electric telescopic rod, for directly driving the second sealing plate 52 to open or close; when the second sealing plate 52 is not opened, the support arm 3 cannot be opened.

[0099] In another optional implementation of this embodiment, the telescopic member 51 is a spring telescopic structure, and the two ends of the telescopic member 51 are elastically connected to the housing 11 and the second sealing plate 52 respectively; the second sealing plate 52 is provided with a slope structure and faces the interior of the housing 11.

[0100] When the support arm 3 switches from the stowed state to the unfolded state for loading, the support arm 3 contacts the inclined surface structure first, so that the second sealing plate 52 can be compressed and contracted adaptively after being compressed, providing an automatic avoidance function for the unfolding of the support arm 3;

[0101] When the support arm 3 switches from the extended state to the retracted state, the support arm 3 rotates and is retracted in the housing 11, and the second sealing plate 52 automatically extends from the inside of the buffer groove 112 under the elastic force of the telescopic member 51 and completely closes the rotating opening 111. Under the action of the same power drive, the rotation and retraction of the support arm 3, the automatic closing of the first sealing plate 4 and the automatic closing of the second sealing plate 52 are synchronously completed.

[0102] In a preferred embodiment of the present invention, the support arm 3 is made of an insulating rubber structure and has anti-slip grooves on its surface.

[0103] The anti-slip performance of the support arm 3 when it is unfolded is increased to ensure the stability of the entire housing 11 when the support arm 3 is unfolded.

[0104] The working principle of the intelligent terminal system with dialect speech recognition provided in this embodiment is as follows:

[0105] like Fig. 9 (a) and Fig.10 As shown in (a), it may be defined that in the initial state, the support arm 3 is in an unfolded state, the playback hole 14 is in an open state, and the rotating mouth 111 is in an open state;

[0106] Please refer to Fig. 9 (a) to Fig. 9 (b) to Fig. 9 (c) When the support arm 3 needs to be folded, the driving member 21 is started, the driving member 21 drives the connecting shaft 221 to rotate, the connecting shaft 221 drives the first bevel gear 222 to rotate, the first bevel gear 222 drives the second bevel gear 223 to rotate, the second bevel gear 223 drives the rotating shaft 224 to rotate, the rotating shaft 224 drives the support arm 3 to rotate counterclockwise, so that the support arm 3 gradually rotates and is stored inside the housing 11, and the support arm 3 switches from the extended state to the retracted state;

[0107] Please refer to Fig.10(a) to Fig.10 (b) to (c), when the connecting shaft 221 rotates, the connecting gear 231 rotates synchronously, the connecting gear 231 drives the connecting tooth plate 232 to move, and the connecting tooth plate 232 drives the first sealing plate 4 to move horizontally, so that the first sealing plate 4 moves and aligns with the range of the playback hole 14, so that the playback hole 14 is automatically closed;

[0108] During the process of the support arm 3 rotating and retracting, the support arm 3 gradually separates from the second sealing plate 52. When the support arm 3 is completely separated, the telescopic member 51 pushes the second sealing plate 52 out of the buffer groove 112 under the action of elastic force, and the second sealing plate 52 is completely closed within the range of the rotating opening 111, so that the rotating opening 111 is automatically closed.

[0109] Finally, under the action of the same power drive, the support arm 3 is stored, the playback hole 14 is closed, and the rotating mouth 111 is closed simultaneously; so as to increase the sealing and protection of the housing 11 at different times and reduce the intrusion of external dust and water stains.

[0110] like Fig. 9 (a) and Fig.10 As shown in (a), the support arm 3 is in a retracted state, the playback hole 14 is in a closed state, and the rotating opening 111 is in a closed state.

[0111] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. An intelligent terminal system with dialect speech recognition, characterized in that: include: Voice collection module, central processing module, data storage module, power supply module, network module and voice broadcast module; The speech collection module includes a speech management unit and a speech recognition unit, wherein the speech management unit is used to construct and update a dialect feature group; The speech recognition unit collects the user's real-time dialect speech through the terminal device; The data storage module is used for storing dialect feature groups; The central processing module performs feature extraction, feature matching and language conversion on the real-time dialect speech. When the user uses dialect input, the central processing module quickly extracts the real-time dialect features and matches the real-time dialect features with the dialect feature group. After the matching is completed, the real-time dialect features are automatically converted into Mandarin features to provide support for the understanding and recognition of the central processing module; The power supply module is used for powering the equipment; The network module is used for network access of the device; The voice broadcast module is used to broadcast the voice content played by the system; The intelligent terminal system with dialect speech recognition is integrated into an intelligent terminal device for speech collection and text conversion; The intelligent terminal device comprises: An all-in-one machine, comprising a housing, a touch screen, a microphone, a playback hole and a camera; a rotating port and a buffer slot are provided on the housing, and the rotating port is connected to the buffer slot; the touch screen, the microphone and the camera are integrated and installed on the housing, and the housing is provided with a playback hole; An adjusting mechanism, wherein the adjusting mechanism comprises a driving member, a connecting assembly and a telescopic assembly, wherein the driving member is fixedly arranged in the housing, the connecting assembly comprises a connecting shaft, a first bevel gear, a second bevel gear and a rotating shaft, the top end of the connecting shaft is fixedly connected to the shaft end of the driving member, the bottom end of the connecting shaft is fixedly connected to the first bevel gear, the second bevel gear is rotatably mounted on the housing through the rotating shaft, the second bevel gear is meshingly connected to the first bevel gear, the telescopic assembly comprises a connecting gear and a connecting tooth plate, the connecting gear is fixedly arranged on the connecting shaft, and the connecting tooth plate is meshingly connected to the connecting gear; A support arm, one end of which is fixedly connected to the rotating shaft, the other end of which passes through the rotating opening and is supported on a table, and the rotation range of the support arm matches the size of the rotating opening; a first sealing plate, the first sealing plate being slidably mounted in the housing, the connecting tooth plate being fixedly connected to the first sealing plate, and the sliding range of the first sealing plate matching the size of the playback hole; A closing device, the closing device comprising a telescopic member and a second sealing plate, the telescopic member is installed in the buffer groove, one end of the second sealing plate is inserted into the buffer groove and fixedly connected to the telescopic end of the telescopic member; Wherein, the rotating port, the buffer groove, the driving member, the connecting assembly, the supporting arm and the closing device are each provided with two groups, and are arranged in a one-to-one correspondence; There are two playback holes, which are symmetrically arranged on both sides of the touch screen; the playback holes, the first sealing plate and the connecting shaft are arranged in a one-to-one correspondence; The telescopic member is a spring telescopic structure, and two ends of the telescopic member are elastically connected to the housing and the second sealing plate respectively; the second sealing plate is provided with an inclined surface structure facing the interior of the housing.

2. The intelligent terminal system with dialect speech recognition according to claim 1, characterized in that: The voice management unit is used for the management personnel to perform input management and manual update management on the dialect feature group stored in the data storage module.

3. The intelligent terminal system with dialect speech recognition according to claim 2, characterized in that: The data storage module is integrated and installed in the server management system, and the server management system provides service support of the dialect feature group for the remote terminal equipment through the network module.

4. The intelligent terminal system with dialect speech recognition according to claim 3, characterized in that: The speech recognition unit is used to collect the real-time speech of the user in real time.

5. The intelligent terminal system with dialect speech recognition according to claim 4, characterized in that: The central processing module includes a speech preprocessing unit, a feature extraction unit and a feature matching unit. The speech preprocessing unit is used to perform noise reduction processing on the collected real-time speech to obtain clear speech; the feature extraction unit is used to extract dialect features in the clear speech; the feature matching unit quickly matches the dialect features with the preset dialect feature group, determines which dialect the dialect features belong to, and provides support for language recognition and text conversion.

6. The intelligent terminal system with dialect speech recognition according to claim 5, characterized in that: The central processing module also includes a database construction unit and a data updating unit. The database construction unit collects the dialect feature group in the big data through the network module, and matches the dialect feature group in the big data with the dialect feature group in the data storage module; determines whether an update is required; When an update is required, the data updating unit automatically and adaptively updates the dialect feature group in the data storage module with the dialect feature group in the big data.

Images