CN207946726U - A kind of all-in-one machine based on speech recognition - Google Patents
A kind of all-in-one machine based on speech recognition Download PDFInfo
- Publication number
- CN207946726U CN207946726U CN201820149364.XU CN201820149364U CN207946726U CN 207946726 U CN207946726 U CN 207946726U CN 201820149364 U CN201820149364 U CN 201820149364U CN 207946726 U CN207946726 U CN 207946726U
- Authority
- CN
- China
- Prior art keywords
- voice
- machine
- voice signal
- operational order
- machine host
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The utility model discloses a kind of all-in-one machine based on speech recognition, the all-in-one machine includes:Integrated machine host, voice collector and display;Wherein, the voice collector is connect with the integrated machine host, and the integrated machine host is sent to for the voice signal that user in real is sent out, and by the voice signal;The one machine host, the voice signal, generates operational order corresponding with the voice signal, and execute corresponding operation according to the operational order for identification;The display is connect with the integrated machine host, for being shown to the operating result after the execution operational order.The utility model embodiment, which solves traditional operation mode, influences operating efficiency, such as the input equipment of the contacts such as touch screen, keyboard and mouse, it is that can not touch or be easy accidentally tactile problem when hand is wet, realizes and operated on all-in-one machine by way of speech recognition.
Description
Technical field
The utility model embodiment is related to all-in-one machine field more particularly to a kind of all-in-one machine based on speech recognition.
Background technology
All-in-one machine is a kind of special purpose computer, the application-specific demand for meeting particular place, such as museum exhibit shop
Displaying all-in-one machine, Scenic spot introduction and map check anesthesia all-in-one machine of all-in-one machine, common query all-in-one machine and hospital etc.
Special applications.
Host, display and various output input devices are integrated together by existing all-in-one machine, internal element height collection
At machine only has a power cord.All-in-one machines many at present realize information input and other using touch screen on all-in-one machine
A series of routine operations, also some all-in-one machines be integrated with keyboard and mouse, pass through dull mouse and click, the operations such as keyboard typing
Carry out routine operation.But these traditional operation modes influence operating efficiency, and touch screen, keyboard and mouse are contacts
Input equipment, be that can not touch or be easy accidentally to touch when hand is wet.
Utility model content
The utility model provides a kind of all-in-one machine based on speech recognition, is grasped on all-in-one machine by voice with realizing
Make.
The utility model embodiment provides a kind of all-in-one machine based on speech recognition, and the all-in-one machine includes:All-in-one machine
Host, voice collector and display;Wherein,
The voice collector is connect with the integrated machine host, for the voice signal that user in real is sent out, and
The voice signal is sent to the integrated machine host;
The one machine host, the voice signal, generates operational order corresponding with the voice signal for identification,
And corresponding operation is executed according to the operational order;
The display connect with the integrated machine host, for the operating result after the execution operational order into
Row display.
Further, the integrated machine host includes:Memory, speech analysis device and main controller;Wherein,
The memory, for storing pre-set phonetic order model;Wherein, the phonetic order model includes each
Correspondence between a voice signal and operational order;
The speech analysis device, for according to voice signal described in the phonetic order model analyzing, the voice to be believed
Number be converted to corresponding operational order;
The main controller, for executing corresponding feature operation in the all-in-one machine according to the operational order.
Further, the voice collector uses microphone array, the microphone array to include at least a Mike
Wind.
Further, the operational order includes:Open application, upper and lower page turning, page zoom-in and zoom-out and page point hit to
Few one kind.
Further, the speech analysis device includes:Voice recognition unit and instruction-determining unit;Wherein,
The voice recognition unit, for by the voice signal with it is pre-set each in the phonetic order model
Voice signal carries out matching identification one by one;
Described instruction determination unit, for when the voice signal identifies successfully, the voice signal to be converted to institute
Predicate sound signal respective operations instruct.
Further, the integrated machine host further includes:
Voice acquisition module obtains for the server belonging to the speech database and needs pre-set each standard speech
Sound signal;
Voice mapping block is established voice and is referred to for setting corresponding operational order to each standard voice signals
It includes the corresponding pass between each standard voice signals and corresponding operational order to enable model, the phonetic order model
System.
Further, the integrated machine host further includes:
Communication module is connect with the voice acquisition module, is used for when starting the speech identifying function of the all-in-one machine,
The voice signal of standard is obtained in server belonging to speech database, and the standard voice signals are sent to institute's predicate
Sound acquisition module.
Further, the integrated machine host further includes:
Voice correction module is used for when executing corresponding operation on the display according to the operational order, right
The corresponding voice signal of the operational order is corrected, and the voice signal after correction is uploaded to belonging to speech database
Server.
Further, the all-in-one machine further includes:Keyboard, mouse and/or loud speaker.
Further, the display uses touch display screen;Between the voice collector and the integrated machine host
Using I2C buses connect.
The utility model embodiment provides a kind of all-in-one machine based on speech recognition, which includes:Integrated owner
Machine, voice collector and display, the voice signal sent out by voice collector user in real, and voice signal is sent out
It send to integrated machine host, then by all-in-one machine recognition of speech signals generation operational order corresponding with voice signal, and according to
Operational order executes corresponding operation, is shown to the operating result after executing operational order finally by display.This reality
Solving traditional operation mode with new embodiment influences the contacts such as operating efficiency, such as touch screen, keyboard and mouse
Input equipment is that can not touch or be easy accidentally tactile problem when hand is wet, realize by way of speech recognition
It is operated on all-in-one machine.
Description of the drawings
Fig. 1 is the structural schematic diagram for the all-in-one machine based on speech recognition that the utility model embodiment one provides;
Fig. 2 is the structural schematic diagram for the all-in-one machine based on speech recognition that the utility model embodiment two provides.
Specific implementation mode
The utility model is described in further detail with reference to the accompanying drawings and examples.It is understood that herein
Described specific embodiment is used only for explaining the utility model, rather than the restriction to the utility model.It further needs exist for
It is bright, it illustrates only for ease of description, in attached drawing and the relevant part of the utility model rather than entire infrastructure.
Embodiment one
Fig. 1 is the structural schematic diagram for the all-in-one machine based on speech recognition that the utility model embodiment one provides, the one
Machine may include:Voice collector 110, integrated machine host 120 and display 130;Wherein,
Voice collector 110 is connect with integrated machine host 120, for the voice signal that user in real is sent out, and will
Voice signal is sent to integrated machine host.
Integrated machine host 120, voice signal, generates operational order corresponding with voice signal, and according to behaviour for identification
Make instruction execution to operate accordingly.
Display 130 is connect with integrated machine host 120, for being shown to the operating result after execution operational order.
In the present embodiment, the voice signal that voice collector 110 can be sent out with user in real.Wherein, voice is adopted
Microphone array may be used in storage 110, and microphone array at least may include a microphone.Optionally, microphone array
May include four microphones put by linear systematic, one lift sound wave beam of each two adjacent two microphones formation, four
A microphone forms 3 pickup wave beams, and each wave beam corresponds to 60 degree of lift sound range, so as to realize out of 180 degree
The voice signal that user sends out is obtained, enhances the sound in beam area to a certain extent, weakens the sound outside wave beam, with enhancing
The signal-to-noise ratio of voice signal.Voice collector 110 can convert voice signals into mould after acquiring the voice signal that user sends out
Quasi- audio signal.
In the present embodiment, after obtaining the voice signal that user sends out by voice collector, it is also necessary to user
The voice signal sent out is identified, and can be carried out to the voice signal that voice collector 110 obtains by integrated machine host 120
Identification, to generate the corresponding operational order of voice signal.Optionally, integrated machine host 120 may include memory 1201, language
Sound resolver 1202 and main controller 1203.Memory 1201 can store pre-set phonetic order model;Wherein, voice refers to
It includes the correspondence between each voice signal and operational order to enable model.
In the present embodiment, speech analysis device 1202 can be believed voice according to phonetic order model analyzing voice signal
Number be converted to corresponding operational order.Refer to operation that is, saving various voice signals in phonetic order model
One-to-one relationship between order, after carrying out voice signal to parse determining voice signal by speech analysis device 1202, just
It is assorted that the one-to-one relationship of the voice signal and operational order that can be preserved according to phonetic order model determines that user wishes to carry out
Operation, and convert the voice signal to the operational order that the voice signal represents.Wherein, can include to beat in voice signal
Open application, the keywords such as upper and lower page turning, page zoom-in and zoom-out and the page are clicked, the corresponding operational order of corresponding voice signal also can
To include the instructions such as opening application, upper and lower page turning, page zoom-in and zoom-out and page click.
Optionally, speech analysis device 1202 can specifically include:Voice recognition unit and instruction-determining unit;Wherein, language
Sound recognition unit, for being compared voice signal and pre-set each voice signal in phonetic order model library one by one
Identification;Instruction-determining unit, for when voice signal identifies successfully, converting voice signals into voice signal respective operations and referring to
It enables.Optionally, voice recognition unit can include audio coding circuit, can be by voice collector by audio coding circuit
The analog audio signal of 110 collected voice signals is converted to digital audio and video signals, follow-up for convenience to carry out voice signal
Identification, digital audio and video signals can also be further converted to text, according to the text signal and phonetic order being converted to
The text message of pre-set each voice signal carries out matching identification one by one in model library.Certainly, voice recognition unit can
The digital audio and video signals being converted to direct basis are carried out with pre-set each voice signal in phonetic order model library
Matching identification one by one.
In the present embodiment, main controller 1203 can execute corresponding feature operation according to operational order in all-in-one machine.
Illustratively, main controller 1203 can respond operational order executed on all-in-one machine open application, upper and lower page turning, page zoom-in and zoom-out and
The feature operation that page point is hit, and real-time display is carried out according to the implementing result of operational order on the display 130.It is optional
, which can be anesthesia all-in-one machine, the voice signal of user be obtained by the voice collector 110 of all-in-one machine, then
The voice signal that user sends out is identified by the integrated machine host 120 of all-in-one machine again, operational order conversion and operation refer to
It enables and executing.Can be executed on all-in-one machine according to user voice signal by above-mentioned module user browse corrective surgery information,
The relevant cases data such as case information, the page that can be between more functions by instruction identification jump choosing, and page command is really
Recognize.Simultaneously real-time display can be carried out on the display 130 according to the implementing result of operational order.
Illustratively, user can send instruction by voice to the all-in-one machine based on speech recognition, and voice collector 110 is adopted
Collect the voice signal, carrying out parsing identification by integrated machine host 120 generates corresponding operational order, is controlled by operational order
Display 130 shows corresponding information interface, in order to which user carries out next step operation.For example, user, which says, " opens case
Information ", anesthesia all-in-one machine opens case information according to voice signal in all-in-one machine, and shows case information circle by display
Face.
The utility model embodiment provides a kind of all-in-one machine based on speech recognition, which includes:Integrated owner
Machine, voice collector and display, the voice signal sent out by voice collector user in real, and voice signal is sent out
It send to integrated machine host, then by all-in-one machine recognition of speech signals generation operational order corresponding with voice signal, and according to
Operational order executes corresponding operation, is shown to the operating result after executing operational order finally by display.This reality
Solving traditional operation mode with new embodiment influences the contacts such as operating efficiency, such as touch screen, keyboard and mouse
Input equipment is that can not touch or be easy accidentally tactile problem when hand is wet, realize by way of speech recognition
It is operated on all-in-one machine.
Embodiment two
Fig. 2 is the structural schematic diagram for the all-in-one machine based on speech recognition that the utility model embodiment two provides, the one
Machine may include:Voice collector 110, integrated machine host 120 and display 130;Wherein, integrated machine host 120 can wrap
It includes:Memory 1201, speech analysis device 1202, main controller 1203 and voice correction module 1204.
Memory 1201, for storing pre-set phonetic order model;Wherein, phonetic order model includes each language
Correspondence between sound signal and operational order.
Speech analysis device 1202, for according to phonetic order model analyzing voice signal, convert voice signals into and its
Corresponding operational order.
Main controller 1203, for executing corresponding feature operation in all-in-one machine according to operational order.
Voice correction module 1204 is used for when executing corresponding operation over the display according to operational order, to operation
It instructs corresponding voice signal to be corrected, and the voice signal after correction is uploaded to the server belonging to speech database.
In the present embodiment, user voice signal is identified by memory 1201 and speech analysis device 1202, it will
Voice signal is converted to corresponding operational order, is then executed accordingly in all-in-one machine according to operational order by main controller 1203
Feature operation.It can also be by voice correction module 1204 to the corresponding voice of operational order while executing operational order
Signal is corrected, and the voice signal after correction is uploaded to the server belonging to speech database.In this case, voice school
Positive module 1204 can learn the voice signal that user sends out, analysis user when sending out voice signal with the voice
Difference between the standard voice signals of signal, the voice signal then sent out to the user are corrected, so that one under user
The secondary voice signal that sends out can quickly identify the voice signal later.Illustratively, user is when sending out voice signal,
The voice signal that may be sent out has a certain difference with standard voice signals, such as in the presence of certain poor between dialect and mandarin
It is different, same voice signal is sent out between different speakers is also likely to be present certain difference.Although having differences still this time
It has been identified successfully by repeatedly identifying, then in this case study school can be carried out by voice correction module 1204
Just, realize that next time, there are can quickly be identified very much when such case.
On the basis of the above embodiments, optionally, integrated machine host 120 can also include:
Voice acquisition module 1205 obtains for the server belonging to the speech database and needs pre-set each mark
Quasi- voice signal;Voice mapping block 1206 establishes voice for setting corresponding operational order to each standard voice signals
Demand model, phonetic order model include the correspondence between each standard voice signals and operational order.
In the present embodiment, voice acquisition module 1205 can be obtained from the server belonging to voice data criteria library and be needed
Pre-set each standard voice signals.Wherein, each cloud signal obtained from speech database may each be some marks
Accurate voice signal.An one-to-one operation is assigned by voice mapping block 1206 to each voice signal obtained to refer to
It enables, and a phonetic order model can be established by the relationship between each voice signal and operational order.In this case, when
User can control the various operating functions of all-in-one machine when using all-in-one machine by voice.
Optionally, integrated machine host 120 further includes:Communication module 1207 is connect with voice acquisition module 1205, is used for
When starting the speech identifying function of all-in-one machine, voice signal is obtained in the server belonging to speech database, and voice is believed
Number it is sent to voice acquisition module 1205.
In the present embodiment, voice data can be pulled at any time from server by communication module 1207, simultaneously also
Voice signal that voice correction module 1204 corrects can be sent to by communication module 1207 to the clothes belonging to speech database
Business device is stored.
On the basis of the above embodiments, optionally, all-in-one machine can also include:It keyboard 140, mouse 150 and/or raises one's voice
Device 160.
In the present embodiment, loud speaker 160 can be stero set, using the Europe 2*8 8W loud speakers, for playing audio
Information reminds user to be operated accordingly by the sound.When gesture identification all-in-one machine is anesthesia all-in-one machine, keyboard 140 can
To use Anaesthesia speciality keyboard.Whether mouse 150 can need to be equipped on all-in-one machine according to the custom of user.
On the basis of the above embodiments, optionally, all-in-one machine can also include power management module and system radiating mould
Block.Optionally, power management module includes power supply, switch panel, display power supply, sound card power supply;System radiating module includes
Water-filled radiator, fan.
On the basis of the above embodiments, optionally, touch display screen may be used in display 130;Voice collector 110
It can be connect using I2C buses between integrated machine host 120.Optionally, integrated machine host 120 may include bluetooth module or
The voice signal that voice collector 110 obtains is sent to one by wireless WIFI module by bluetooth module or wireless WIFI module
Body machine host 120.
On the basis of the above embodiments, optionally, all-in-one machine can also include the radio frequency mould being connect with integrated machine host
Block realizes the wireless connection between the various equipment with net, the work of radio-frequency module for constituting radio frequency network in domestic environment
Working frequency is 433MHZ.
In the present embodiment, all-in-one machine will be combined in gesture identification technology with integrated machine host, and realizing all-in-one machine only needs to use
Non-contact voice can complete interactive operation, as selection menu content is brandished in left and right, target is chosen in finger rotation:The enquiry machine
Operation can largely be simplified, reduce cost, it is only necessary to common LCD display, it is no longer necessary to touch tablet beautifies interface,
Virtual functions key need not be placed on interface, can also reduce the probability of cross-infection:Application range is boundless, mainly has
The inquiry of public information, such as telecommunication bureau, the tax bureau, bank, electric power department service inquiry:The information inquiry in city street corner;
In addition, also can be widely used to enterprise's office, Industry Control, military commanding, electronic game, choosing song or selecting dish, multimedia teaching, room
Real estate presell etc..Such as typical anesthesia all-in-one machine, museum exhibit shop displaying all-in-one machine, Scenic spot introduction and map are checked
The special applications such as all-in-one machine, common query all-in-one machine.
The utility model embodiment provides a kind of all-in-one machine based on speech recognition, which includes:Integrated owner
Machine, voice collector and display, the voice signal sent out by voice collector user in real, and voice signal is sent out
It send to integrated machine host, then by all-in-one machine recognition of speech signals generation operational order corresponding with voice signal, and according to
Operational order executes corresponding operation, is shown to the operating result after executing operational order finally by display.This reality
Solving traditional operation mode with new embodiment influences the contacts such as operating efficiency, such as touch screen, keyboard and mouse
Input equipment is that can not touch or be easy accidentally tactile problem when hand is wet, realize by way of speech recognition
It is operated on all-in-one machine.
Note that above are only the preferred embodiment and institute's application technology principle of the utility model.Those skilled in the art's meeting
Understand, the utility model is not limited to specific embodiment here, can carry out for a person skilled in the art various apparent
Change, readjust and substitutes without departing from the scope of protection of the utility model.Therefore, although by above example to this
Utility model is described in further detail, but the utility model is not limited only to above example, is not departing from this reality
Can also include other more equivalent embodiments in the case of with novel design, and the scope of the utility model is by appended power
Sharp claimed range determines.
Claims (7)
1. a kind of all-in-one machine based on speech recognition, which is characterized in that the all-in-one machine includes:Integrated machine host, voice collecting
Device and display;Wherein,
The voice collector is connect with the integrated machine host, for the voice signal that user in real is sent out, and by institute
Predicate sound signal is sent to the integrated machine host;
The one machine host, the voice signal, generates operational order corresponding with the voice signal, and root for identification
Corresponding operation is executed according to the operational order;
The display is connect with the integrated machine host, for being shown to the operating result after the execution operational order
Show;
The voice collector uses microphone array, and the microphone array includes four Mikes put by linear systematic
Wind forms a lift sound wave beam per two adjacent microphones;The display uses touch display screen;The voice collector
I is used between the integrated machine host2C buses connect.
2. all-in-one machine according to claim 1, which is characterized in that it is described one machine host include:Memory, speech analysis
Device and main controller;Wherein,
The memory, for storing pre-set phonetic order model;Wherein, the phonetic order model includes each language
Correspondence between sound signal and operational order;
The speech analysis device, for according to voice signal described in the phonetic order model analyzing, the voice signal to be turned
It is changed to corresponding operational order;
The main controller, for executing corresponding feature operation in the all-in-one machine according to the operational order.
3. all-in-one machine according to claim 1, which is characterized in that the operational order includes:It opens application, turn over up and down
At least one of page, page zoom-in and zoom-out and page click.
4. all-in-one machine according to claim 2, which is characterized in that the speech analysis device includes:Voice recognition unit and
Instruction-determining unit;Wherein,
The voice recognition unit is used for pre-set each voice in the voice signal and the phonetic order model
Signal carries out matching identification one by one;
Described instruction determination unit, for when the voice signal identifies successfully, the voice signal to be converted to institute's predicate
Sound signal respective operations instruct.
5. all-in-one machine according to claim 1, which is characterized in that it is described one machine host further include:
Voice acquisition module obtains for the server belonging to the speech database and needs pre-set each standard speech message
Number;
Voice mapping block establishes phonetic order mould for setting corresponding operational order to each standard voice signals
Type, the phonetic order model include the correspondence between each standard voice signals and corresponding operational order.
6. all-in-one machine according to claim 5, which is characterized in that it is described one machine host further include:
Communication module is connect with the voice acquisition module, is used for when starting the speech identifying function of the all-in-one machine, in language
The voice signal of standard is obtained in server belonging to sound database, and the standard voice signals are sent to the voice and are obtained
Modulus block.
7. all-in-one machine according to claim 1, which is characterized in that the all-in-one machine further includes:It keyboard, mouse and/or raises
Sound device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201820149364.XU CN207946726U (en) | 2018-01-29 | 2018-01-29 | A kind of all-in-one machine based on speech recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201820149364.XU CN207946726U (en) | 2018-01-29 | 2018-01-29 | A kind of all-in-one machine based on speech recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN207946726U true CN207946726U (en) | 2018-10-09 |
Family
ID=63702615
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201820149364.XU Active CN207946726U (en) | 2018-01-29 | 2018-01-29 | A kind of all-in-one machine based on speech recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN207946726U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111968638A (en) * | 2020-08-14 | 2020-11-20 | 上海茂声智能科技有限公司 | Method, system, equipment and storage medium for voice control display terminal |
-
2018
- 2018-01-29 CN CN201820149364.XU patent/CN207946726U/en active Active
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111968638A (en) * | 2020-08-14 | 2020-11-20 | 上海茂声智能科技有限公司 | Method, system, equipment and storage medium for voice control display terminal |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11282519B2 (en) | Voice interaction method, device and computer readable storage medium | |
CN109948046B (en) | Message pushing method and device, mobile terminal and storage medium | |
JP6440513B2 (en) | Information providing method and device control method using voice recognition function | |
US20190025950A1 (en) | User interface apparatus and method for user terminal | |
CN1790326B (en) | System for synchronizing natural language input element and graphical user interface | |
CN106202204A (en) | The lookup method of a kind of voice document and mobile terminal | |
WO2018072543A1 (en) | Model generation method, speech synthesis method and apparatus | |
CN104536711A (en) | Control method of terminal display | |
CN107766455B (en) | Lyric pickup method, device and computer readable storage medium | |
CN101311896A (en) | Computer system possessing multipurpose desktop | |
CN207946726U (en) | A kind of all-in-one machine based on speech recognition | |
CN116070114A (en) | Data set construction method and device, electronic equipment and storage medium | |
Ruminski et al. | Interactions with recognized objects | |
CN106095128B (en) | Character input method of mobile terminal and mobile terminal | |
CN111142993A (en) | Information acquisition method, terminal and computer storage medium | |
CN109597996A (en) | A kind of semanteme analytic method, device, equipment and medium | |
CN112578967B (en) | Chart information reading method and mobile terminal | |
CN112732379B (en) | Method for running application program on intelligent terminal, terminal and storage medium | |
CN112817582A (en) | Code processing method and device, computer equipment and storage medium | |
CN207946727U (en) | A kind of all-in-one machine based on gesture identification | |
CN110175063B (en) | Operation assisting method, device, mobile terminal and storage medium | |
CN111601367A (en) | Network connection control method and system based on voiceprint recognition and mobile terminal | |
CN105677078A (en) | Mobile terminal and shortcut key customizing method thereof | |
CN110493447A (en) | A kind of message treatment method and relevant device | |
WO2022121778A1 (en) | Sound information processing method and apparatus, and computer storage medium and electronic device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |