WO2017156893A1 - 语音控制方法及智能电视 - Google Patents

语音控制方法及智能电视 Download PDF

Info

Publication number
WO2017156893A1
WO2017156893A1 PCT/CN2016/085157 CN2016085157W WO2017156893A1 WO 2017156893 A1 WO2017156893 A1 WO 2017156893A1 CN 2016085157 W CN2016085157 W CN 2016085157W WO 2017156893 A1 WO2017156893 A1 WO 2017156893A1
Authority
WO
WIPO (PCT)
Prior art keywords
semantic
information
semantic information
text
keyword
Prior art date
Application number
PCT/CN2016/085157
Other languages
English (en)
French (fr)
Inventor
廖洪涛
胡佳文
王云华
Original Assignee
深圳Tcl数字技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳Tcl数字技术有限公司 filed Critical 深圳Tcl数字技术有限公司
Publication of WO2017156893A1 publication Critical patent/WO2017156893A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/30—Driver circuits
    • H05B45/32—Pulse-control circuits
    • H05B45/325—Pulse-width modulation [PWM]
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/30—Driver circuits
    • H05B45/395—Linear regulators
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
    • Y02B20/30—Semiconductor lamps, e.g. solid state lamps [SSL] light emitting diodes [LED] or organic LED [OLED]

Definitions

  • the present invention relates to the field of smart televisions, and in particular, to a voice control method and a smart television.
  • TV sets are also moving towards an intelligent trend.
  • smart TVs also have network functions that enable cross-platform search between TVs, networks and programs.
  • Smart TV is becoming the third kind of information access terminal after computers and mobile phones. Users can access the information they need through smart TV.
  • the smart television has a voice control function.
  • the voice feature library is pre-stored in the smart television. After receiving the input voice signal, the voice feature of the voice signal is extracted and matched in the preset voice feature database. The voice feature generates a corresponding operation instruction according to the matching result of the voice feature to implement voice control.
  • the matching can only be performed according to the preset voice feature library, and the content of the voice input by the user cannot be truly understood, and the problem of matching errors, the inability to accurately identify, and the user experience are reduced.
  • the main purpose of the present invention is to provide a voice control method and a smart television, which are intended to solve the problem of easy identification of smart televisions controlled by voice recognition in the prior art, and to understand voice content using natural semantics. A problem that causes a poor user experience.
  • the present invention provides a voice control method, including:
  • the smart television receives the text information sent by the voice input device, and the text information is obtained by the voice input device performing voice recognition on the voice signal input by the user;
  • the smart TV performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information;
  • the smart television controls the smart television according to the semantic information
  • the smart TV performs semantic analysis on the text information according to a preset standard code comparison table, including:
  • the semantics of each of the text characters are determined according to a standard code comparison table preset in a language to which each of the text characters belongs.
  • the smart TV controls the smart TV according to the semantic information, including:
  • the semantic database includes the semantic information
  • obtaining operation data corresponding to the semantic information from the semantic database and controlling the smart television according to the operation data, where the semantic database includes semantics The correspondence between information and operational data.
  • the present invention provides a voice control method, including:
  • the smart television receives the text information sent by the voice input device, and the text information is obtained by the voice input device performing voice recognition on the voice signal input by the user;
  • the smart TV performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information;
  • the smart television controls the smart television according to the semantic information.
  • the present invention provides a smart television, including:
  • a receiving module configured to receive text information sent by the voice input device, where the text information is obtained by the voice input device performing voice recognition on a voice signal input by the user;
  • An analysis module configured to perform semantic analysis on the text information according to a standard coding comparison table of a preset language, to obtain semantic information of the text information;
  • control module configured to control the smart TV according to the semantic information.
  • the smart television after acquiring the text information of the voice signal input by the user, the smart television performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information, and The smart TV is controlled according to the semantic information, and the semantic analysis of the text information of the voice signal enables the semantics of the voice signal input by the user to be effectively recognized, the natural semantics of the voice signal is understood, and the recognition accuracy is increased, thereby enabling Better interact with users to improve the user experience.
  • FIG. 1 is a schematic flow chart of a voice control method according to a first embodiment of the present invention
  • FIG. 2 is a schematic flow chart showing the steps of the semantic analysis of the text information by the smart TV of FIG. 1 according to a preset standard code comparison table;
  • FIG. 3 is a schematic flowchart diagram of a voice control method according to a second embodiment of the present invention.
  • FIG. 4 is a schematic diagram of functional modules of a smart television according to a third embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a refinement function module of the analysis module 402 of the smart TV in the third embodiment of FIG. 4;
  • FIG. 6 is a schematic diagram of functional modules of a smart television according to a fourth embodiment of the present invention.
  • the smart television after acquiring the text information of the voice signal input by the user, the smart television performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information, and The smart TV is controlled according to the semantic information, so that the smart TV can realize the natural semantics of the voice signal, increase the recognition accuracy, and enable better interaction with the user and improve the user experience.
  • a voice control system in the embodiment of the present invention includes a voice input device and a smart TV.
  • the voice input device includes a voice input function and a voice recognition function, wherein the voice input function can be implemented by a microphone disposed on the voice input device, and the voice input device can use the microphone to collect the voice of the user and generate voice signal.
  • the voice recognition function of the voice input device can be implemented by a voice recognition module in the voice input device. After the voice signal is generated, the microphone transmits the voice signal to the voice recognition module, and the voice recognition module identifies the voice signal, and obtains the voice signal. Text information of the voice signal.
  • the voice input device may be a remote controller or other mobile terminal, and the other mobile terminals may be smart terminal devices such as a smart phone, a tablet computer, and a palmtop computer.
  • the method for recognizing the voice signal input by the voice recognition module may be: performing PCM coding on the voice signal to obtain a PCM code stream, and using the obtained PCM code stream and the PCM code stream and text characters preset in the voice input device. Comparing the correspondence between the character codes of the text characters, determining the character codes of the text characters and the text characters corresponding to each PCM code stream, to obtain the text information of the voice signal, the text information including the text characters in the voice signal and The character encoding of the text character.
  • the PCM stream of the voice signal is: 0x0100 (Liu), 0x0200 (German), 0x0300 (Hua), 0x0400 (New), 0x0500 (Smell), in parentheses
  • the text character corresponding to the stream is: 0x0100 (Liu), 0x0200 (German), 0x0300 (Hua), 0x0400 (New), 0x0500 (Smell), in parentheses The text character corresponding to the stream.
  • the smart TV is a terminal equipped with an operating system, can freely install and uninstall software programs, has functions of video, entertainment, games, etc., and can implement network functions by wired or wireless means.
  • both the voice input device and the smart TV have an infinite communication module, and the voice input device and the smart TV communicate wirelessly through respective wireless communication modules, thereby establishing a wireless voice channel between the smart TV and the voice input device.
  • the wireless communication module may be a WIFI module, a Bluetooth module, or an infinite USB module, and the like, which is not limited herein.
  • a voice control method includes:
  • Step 101 The smart television receives the text information sent by the voice input device, and the text information is obtained by the voice input device performing voice recognition on the voice signal input by the user;
  • the voice input device after inputting the voice signal input by the user and performing voice recognition on the voice signal to obtain the text information of the voice signal, the voice input device sends the text information of the voice signal to the smart television, wherein the smart television
  • the voice input device and the voice input device can transmit the text information of the voice signal through the wireless voice channel described above.
  • the smart television will receive the text information of the voice signal, and the text information includes the character code of the text character and the text character in the voice signal.
  • Step 102 The smart TV performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information.
  • the smart television after receiving the text information of the voice signal, performs semantic analysis on the text information according to a standard code comparison table of a preset language to obtain semantic information of the text information.
  • a standard code comparison table of a plurality of different languages may be preset in the smart TV.
  • a Unicode code comparison table may be preset in Chinese
  • a US standard information exchange code ASCII
  • ASCII US standard information exchange code
  • English American Standard Code for Information Interchange
  • Unicode is a character encoding used on a computer, which sets a uniform and unique binary encoding for each character in each language, and preferably, UTF can be used for the standard encoding comparison table used for Chinese. -8 (8-bit Unicode Transformation Format), UTF-8 is a variable length character encoding for Unicode, which is one of the implementations of Unicode.
  • ASCII is a computer coding system based on the Latin alphabet, mainly used to display modern English and other Western European languages.
  • Step 103 The smart TV controls the smart TV according to the semantic information.
  • the smart TV after the smart TV parses the semantic information, the smart TV can control the smart television according to the semantic information.
  • the smart television after acquiring the text information of the voice signal input by the user, the smart television performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information, and The smart TV is controlled according to the semantic information, so that the semantics of the voice signal input by the user can be effectively recognized, the natural semantics of the voice signal is understood, the recognition accuracy is increased, and the user can interact better and improve the user experience.
  • FIG. 2 is a flowchart of the step of refining the semantic analysis of the smart TV according to the preset standard code according to the preset standard code in the first embodiment of the present invention, including:
  • Step 201 Determine, according to a character encoding size of each text character in the text information and a character size area of a preset language, a language to which each text character belongs;
  • Step 202 Determine the semantics of each text character according to a standard code comparison table preset in a language to which each text character belongs.
  • the smart television can determine the language to which each text character belongs according to the character encoding size of each text character in the text information and the character size area of the preset language, and can also belong to each text character.
  • the language's pre-set standard code comparison table determines the semantics of each text character.
  • the language to which each text character belongs is determined, specifically: the smart television sequentially determines whether the character encoding of the text character in the text information is greater than 255, when When the character code of a text character is greater than 255, it is determined that the text character belongs to a Chinese character. When the character code of a text character is less than 255, it is determined that the text character belongs to an English character. Because the character size of Chinese characters in Unicode encoding rules is greater than 255, the size of English characters in ASCII is less than 255. Among them, English characters include English words, letters and Arabic numerals.
  • the semantics of each text character is determined according to the standard coding comparison table of the language to which each text character belongs in step 202. Specifically, the smart television sequentially traverses the text characters in the text information, and the traversed text characters are Chinese characters.
  • the text character is subjected to Chinese differential filtering fuzzy matching calculation using a preset Unicode encoding comparison table, the semantics of the text character is determined, and when the traversed text character belongs to an English character, the preset ASCII encoding comparison table is used.
  • the text characters are subjected to English differential screening fuzzy matching calculation to determine the meaning of the text characters. Through the above traversal and matching process, the semantic information of the text information can be determined.
  • the Chinese differential filtering fuzzy matching calculation is performed on the characters by using a preset Unicode encoding comparison table to determine the semantics of the character.
  • the character encoding of the text character is Unicode variable.
  • Length character encoding which sequentially calculates the difference between the variable length character encoding of the text character to be matched and the variable length character encoding of all Chinese characters in the preset Unicode encoding comparison table, when determining a preset Unicode encoding comparison If the difference between the variable length character encoding of the Nth Chinese character in the table and the variable length character encoding of the text character to be matched is 0, it is determined that the text character to be matched is the Nth Chinese character.
  • the semantics of the Nth Chinese character is taken as the semantics of the text character to be matched. And stopping the Chinese differential filtering fuzzy matching calculation for the text characters to be matched. For example, if the variable length character encoding of the text character to be matched is 0x5218, the variable length character encoding is sequentially subtracted from the variable character encoding of the Chinese character in the Unicode encoding comparison table, when 0x5218 and the 100th Chinese are determined. When the difference between the variable length character code 0x5218 of the character is 0, the semantics corresponding to the 100th Chinese character (the semantic is: Liu) is determined as the semantics of the text character to be matched.
  • the character of the character to be matched is determined by using a pre-set ASCII code comparison table to perform an English differential filtering fuzzy matching calculation on the character to determine the meaning of the character.
  • the ASCII code sequentially calculates the difference between the ASCII code of the text character to be matched and the ASCII code of all the English characters in the preset ASCII code comparison table, and determines the Mth English in the preset ASCII code comparison table. If the difference between the ASCII code of the character and the ASCII character of the character to be matched is 0, it is determined that the character to be matched is the Nth Chinese character, and the semantics of the Nth Chinese character is used as the text character to be matched. Semantics.
  • the ASCII code of the text character to be matched is 49
  • the ASCII code 49 is sequentially subtracted from the ASCII code of the English character in the ASCII code, and the difference between the ASCII code 49 and the ASCII code 49 of the 50th English character is determined.
  • the semantics corresponding to the 50th English character is the semantics of the text character to be matched.
  • the language to which each text character in the text information belongs is determined first, and then the fuzzy matching calculation is performed on each text character in turn as an example to describe the process of identifying the semantic information.
  • it is also possible to directly traverse the text characters in the text information first determine whether the character encoding of the traversed text characters is greater than 255, and perform fuzzy matching calculation according to the judgment result to determine the semantic information of the text information.
  • the language pre-set by the smart TV is described in Chinese and English.
  • the voice signal input by the voice input device is used, the language is not currently used by the smart TV.
  • the smart TV can first convert the text information of the speech signal after the speech recognition into the text information that can be recognized according to the language preset by the smart TV. For example, if the user inputs voice in Monaco, the voice input device can convert the recognized text information of Monaco into Chinese and/or English text information after performing voice recognition.
  • the standard code comparison table of multiple languages can be preset in the smart TVs used in different environments and different countries, the user can select the language used by the smart TV, and the smart TV can use the The standard code comparison table of the language used performs the above steps 201 and 202 instead of the standard code comparison table of the language set in advance.
  • the smart code can determine the semantics of each text character in the text information of the voice signal based on the standard code comparison table of different languages by presetting the standard code comparison table used in different languages in the smart TV. Effectively complete the recognition of semantics. And the accuracy of semantic recognition is high, which can give users a better experience.
  • a voice control method includes:
  • Step 301 The smart television receives the text information sent by the voice input device, where the text information is obtained by the voice input device performing voice recognition on the voice signal input by the user;
  • step 301 the content described in step 301 is similar to the content described in step 101 in the first embodiment shown in FIG. 1, and details are not described herein.
  • Step 302 The smart TV performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain semantic information of the text information.
  • step 302 the content described in step 302 is similar to the content described in step 102 in the first embodiment shown in FIG. 1, and the refinement process of step 101 shown in FIG. 2 is also the step 302 in the present invention. The process is not described here.
  • Step 303 determining whether the semantic data is currently stored in the semantic database; when the semantic database contains semantic information, performing step 304, when the semantic information library does not contain semantic information, then performing step 305;
  • the smart television determines whether the semantic information is included in the saved semantic database.
  • the semantic database includes a correspondence between the semantic information and the operation data.
  • the operation data corresponding to the volume is a volume volume up command.
  • the voice information is a Jay Chou movie
  • the operation data corresponding to the Jay Chou movie is a picture and/or text information of a network link address corresponding to the Jay Chou movie displayed on the display interface of the smart TV. Therefore, the operational data corresponding to different semantic information in the semantic database is also different.
  • Step 304 Obtain operation data corresponding to the semantic information from the semantic database, and control the smart television according to the operation data, where the semantic database includes a correspondence relationship between the semantic information and the operation data;
  • the smart television acquires operation data corresponding to the semantic information from the semantic database, and controls the smart television according to the operation data, for example, if the semantic information If the volume is lowered, the operation data corresponding to the reduced volume is a volume down command, and the smart TV will lower the volume of the preset size according to the volume down command.
  • the voice information is an Andy Lau movie
  • the operation data corresponding to the Andy Lau movie is to display the picture and/or text information of the network link address corresponding to the Andy Lau movie on the display interface of the smart TV, and the smart TV will read from the voice database.
  • the network link address corresponding to the Andy Lau movie is taken, and the picture and/or text information of the network link address is displayed on the display interface of the smart TV, so that the user can select a specific movie for viewing.
  • Step 305 Perform semantic word segmentation according to semantics to obtain words included in the semantic information
  • Step 306 Match the words included in the semantic information with the words in the set of words corresponding to different classification entries in the preset media data, determine the classification entries matching each word in the semantic information, and the matching classification entries. Number of
  • Step 307 A word that matches the number of matching classification items that meets a preset numerical condition is used as a keyword;
  • Step 308 Based on the network crawler mechanism, obtain a network link address corresponding to the keyword under the category entry according to the keyword and the keyword matching category entry, and the network link address is the number of clicks of the keyword under the category entry satisfying the preset number of clicks. the address of;
  • the smart television processes the semantic information according to the semantics, and obtains the words contained in the semantic information, for example, for the semantic information “Andy News”, the smart TV follows Semantic division can get the words “Andy Lau” and "news.”
  • the smart television matches the words included in the semantic information with pre-set media data, wherein the media data includes different classification entries, and each category entry corresponds to a word Collections, for example: media data contains classification items: news, movies, TV series, variety, music, etc., under the category of variety, the words collection contains "Andy Lau, Jay Chou, where is Dad." .
  • the smart TV matches the words with the words in the set of words corresponding to each of the category entries in the media data to determine the category entries that the word matches, and the number of matching category entries .
  • the matching category entries include: movie, star, variety, music, and the number of matching category entries is four.
  • the smart television after obtaining the number of classification entries matched by the words in the semantic information in the different classification entries of the media data, the smart television sets the number of the matched classification entries to satisfy the pre-set numerical condition.
  • a keyword wherein the number of matched categorical entries satisfying a pre-set numerical condition refers to a word that matches the largest number of non-categorical entries, or a number of matching categorical entries ranked in the top N, wherein N is a positive integer.
  • the number of matching classification entries of the word "Andy Lau" contained in the word collection is four, and the number of classification entries matching the word "news" is one, and the word with the largest number of matching classification items is determined.
  • Liu Dehua is a key word.
  • the smart TV after obtaining the keyword, acquires a network link address corresponding to the different classification items matched by the keyword according to the keyword and the matching category entry of the keyword based on the network crawler mechanism.
  • the network link address is an address that satisfies a preset number of clicks based on the keyword's number of clicks under the different category entries that match the keyword.
  • the address that satisfies the preset number of clicks may be an address in which the number of clicks is ranked in the top N, and the N is a positive integer.
  • the smart TV obtains the network link address corresponding to the different classification items matched by the keyword based on the web crawler mechanism, and specifically: the smart TV provides the category entry of the keyword and the keyword matching to the web crawler, and the web crawler will Based on the keyword and the classification item matched by the keyword, the network connection address of the keyword under the different classification entries matched with the keyword is obtained from the cloud server, and the click of each network link address under the different classification entries is counted. Number of times, and sorting the number of clicks of the network link address under each of the above-mentioned category entries, and sorting the N network address addresses of the top N under the category entry as the network link address corresponding to the keyword under the category entry .
  • the keyword is “Andy Lau”
  • the classified items matching the keyword include: movie, TV, music, variety
  • the web crawler will use the keyword “Andy Lau” and its classification items to capture the keyword “Andy Lau”
  • the network link addresses under the category entries of movies, TV, music, variety shows, etc., and the network link addresses of different category entries are counted and sorted, and the four network links with the top 4 clicks in the movie category entry are determined.
  • the address is the corresponding network link address of the keyword "Andy Lau” under the movie classification entry, and the keyword "Andy Lau” can also determine the corresponding network link address under the category of TV, music, variety and other categories.
  • Step 309 Display a picture and/or a text corresponding to the network link address corresponding to the keyword under the category entry on the display interface of the smart TV.
  • the smart TV after obtaining the network link address corresponding to the keyword under which the keyword matches, the smart TV selects the picture and/or the picture corresponding to the network link address corresponding to the matching category entry.
  • the text is displayed on the display program of the smart TV, enabling the user to select the displayed content.
  • step 310 namely:
  • Step 310 The network link address is used as operation data of the semantic information, and the correspondence between the semantic information and the network link address is saved in the semantic database.
  • the smart TV after acquiring the network link address corresponding to the keyword in the semantic information, uses the network link address as the operation data of the semantic information, and associates the semantic information with the network link address.
  • the relationship is saved in the semantic database.
  • the smart TV may perform step 309 and step 310 simultaneously after performing step 308, or may continue to perform step 309 after performing step 308, and then perform step 310 after performing step 309, or may perform After step 308, step 310 is performed first and then step 309 is performed, that is, the order of execution of step 309 and step 310 is not limited.
  • the smart TV will periodically clean up the semantic database according to a preset policy, and the cleaning period can be preset, for example, one can be set. It can be cleaned once a week, or it can be set up once a month.
  • the specific cleaning cycle is not limited.
  • the semantic information in the semantic database may also be classified, for example, may be classified into semantic information for volume, channel, switch, etc. of the smart television, and content displayed for the smart television. Semantic information.
  • the smart TV can clean up the semantic database according to a preset cleaning policy.
  • the semantic database can be completely cleared, or a certain type of semantic information and corresponding operation data in the semantic database can be cleaned up, for example, for example. It is possible to clean only the semantic information of the displayed content of the smart TV.
  • the smart television adds a value of the semantic information to the semantic database if it is determined that the semantic information is included in the semantic database after each judgment determines whether the semantic information is included in the semantic database.
  • the smart television adds a value of the semantic information to the semantic database if it is determined that the semantic information is included in the semantic database after each judgment determines whether the semantic information is included in the semantic database.
  • the semantic database when the semantic database is cleaned up, it can also be cleaned according to the count value of the semantic information in the semantic database. For example, the semantic information whose count value is smaller than a certain value and the corresponding operation data can be cleaned up.
  • the time point of adding the count value of the semantic information to 1 may be recorded, so that the closest time point distance of the record can be recorded by the semantic information when the semantic database is cleaned up.
  • the semantic database may be preset, so that after the semantic information of the text information is analyzed and analyzed, if the semantic information is found in the semantic database, the semantic information is obtained from the semantic database.
  • the operation data if the semantic information is not included in the semantic database, retrieves the network link address corresponding to the keyword included in the semantic information from the network to obtain the operation data of the semantic information. Because the analysis and recognition of semantic information is high and not easy to make mistakes, the operations performed on the smart TV are all operations that the user wants, which can effectively realize the understanding of the natural semantics of the user by the smart TV and improve the user experience.
  • FIG. 4 is a schematic diagram of functional modules of a smart television according to a third embodiment of the present invention, including:
  • the receiving module 401 is configured to receive text information sent by the voice input device, where the text information is obtained by the voice input device performing voice recognition on the voice signal input by the user;
  • the analyzing module 402 is configured to perform semantic analysis on the text information according to a standard encoding comparison table of a preset language to obtain semantic information of the text information;
  • the control module 403 is configured to control the smart TV according to the semantic information.
  • the smart television after acquiring the text information of the voice signal input by the user, the smart television performs semantic analysis on the text information according to a standard coding comparison table of a preset language to obtain voice information of the text information, and according to the The semantic information controls the smart TV, which can effectively recognize the semantics of the voice signal input by the user, realize the understanding of the natural semantics of the voice signal, increase the recognition accuracy, and enable better interaction with the user and improve the user experience.
  • FIG. 5 it is a refinement function module of the analysis module 402 in the third embodiment shown in FIG. 4, which includes:
  • a first determining module 501 configured to determine, according to a character encoding size of each text character in the text information and a character size area of a preset language, a language to which each text character belongs;
  • the semantic analysis module 502 is configured to determine the semantics of each of the text characters according to a standard encoding comparison table preset in a language to which each text character belongs.
  • the first determining module 501 is specifically configured to:
  • each refinement function module in the analysis module corresponds to the operation in each method step in FIG. 2.
  • the specific content of these operation steps has been described in detail in the foregoing, and therefore will not be described here.
  • the smart code can determine the semantics of each text character in the text information of the voice signal based on the standard code comparison table of the different language by presetting the standard code comparison table used in different languages in the smart TV. , effectively complete the recognition of semantics. And the accuracy of semantic recognition is high, which can give users a better experience.
  • FIG. 6 is a schematic diagram of a functional module of a smart television according to a fourth embodiment of the present invention, including: a receiving module 401, an analyzing module 402, and a control module 403 in the third embodiment shown in FIG.
  • the content described in the embodiment shown in FIG. 4 is similar and will not be described herein.
  • the control module 403 includes:
  • the determining module 601 is configured to determine whether semantic information is included in the currently saved semantic database
  • the first processing module 602 is configured to: when the semantic information is included in the semantic database, obtain operation data corresponding to the semantic information from the semantic database, and control the smart television according to the operation data, where the semantic database includes The correspondence between semantic information and operational data.
  • the control module 403 further includes:
  • the word segmentation module 603 is configured to process semantic information according to semantics when the semantic database does not include semantic information, and obtain words included in the semantic information;
  • the matching module 604 is configured to match the words included in the semantic information with the words in the set of words corresponding to different classification entries in the preset media data, determine the category entries matching each word in the semantic information, and match The number of classification entries;
  • the keyword determining module 605 is configured to use, as a keyword, a word that matches the number of matching classification items that meets a preset numerical condition;
  • the address obtaining module 606 is configured to obtain, according to the network crawler mechanism, the network link address corresponding to the keyword under the category entry according to the keyword and the keyword matching category entry, and the network link address is the keyword.
  • the display module 607 is configured to display a picture and/or a text corresponding to the network link address corresponding to the keyword under the category entry on the display interface of the smart TV.
  • the smart television further includes:
  • the saving module 608 is configured to use the network link address as the operation data of the semantic information, and save the correspondence between the semantic information and the network link address in the semantic database.
  • the cleaning module 609 is configured to clean up the semantic database according to a preset cleaning policy when the semantic database uncleaning time reaches a preset duration.
  • each module in the smart TV in this embodiment corresponds to an operation in each method step in FIG.
  • the specific content of these operation steps has been described in detail in the foregoing, and therefore will not be described here.
  • the cleaning module 609 is connected to the saving module 608, and the function of the cleaning module 609 can be executed after the function of the saving module 608.
  • the function of the cleaning module 609 is It can also be executed before or after the function execution of any module other than the saving module 608 in the fourth embodiment shown in FIG. 6, that is, the connection relationship between the cleaning module 609 and other modules is not limited, as long as the semantic database is not The cleaning time reaches the preset time, and the startup cleaning module 609 can clean the semantic database according to the preset cleaning policy.
  • the semantic database may be preset, so that after the semantic information of the text information is analyzed and analyzed, if the semantic information is found in the semantic database, the semantic information is obtained from the semantic database.
  • the operation data if the semantic information is not included in the semantic database, retrieves the network link address corresponding to the keyword included in the semantic information from the network to obtain the operation data of the semantic information. Because the analysis and recognition of semantic information is high and not easy to make mistakes, the operations performed on the smart TV are all operations that the user wants, which can effectively realize the understanding of the natural semantics of the user by the smart TV and improve the user experience.
  • the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better.
  • Implementation Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
  • the optical disc includes a number of instructions for causing a terminal device (which may be a cell phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods of various embodiments of the present invention.

Landscapes

  • Circuit Arrangement For Electric Light Sources In General (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • User Interface Of Digital Computer (AREA)
  • Dc-Dc Converters (AREA)

Abstract

本发明公开了一种语音控制方法，该语音控制方法包括：智能电视接收语音输入设备发送的文本信息，该文本信息为语音输入设备对用户输入的语音信号进行语音识别得到的，且智能电视将根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，得到文本信息的语义信息，并根据该语义信息对智能电视进行控制。本发明还公开了一种智能电视，能够有效识别用户输入的语音信号的语义，实现对语音信号的自然语义的理解，增加识别准确率，使得能够与用户更好地进行互动，改善用户体验。

Description

语音控制方法及智能电视
技术领域
本发明涉及智能电视领域，尤其涉及一种语音控制方法及智能电视。
背景技术
随着科技的发展，电视机也朝着智能化的趋势发展。智能电视除具有传统的视频、游戏等功能外，还具有网络功能，能够实现电视、网络和程序之间的跨平台搜索。智能电视正在成为继计算机、手机之后的第三种信息访问终端，用户可通过智能电视访问自己需要的信息。
目前，智能电视具有语音控制功能，具体的：智能电视中预先存储了语音特征库，在接收到输入的语音信号之后，提取该语音信号的语音特征，并在该预先设置的语音特征库中匹配该语音特征，并根据语音特征的匹配结果生成对应的操作指令，以实现语音控制。
然而，上述的语音控制过程中，只能根据预先设置的语音特征库进行匹配，无法真正做到理解用户输入的语音的内容，且还容易出现匹配出错，无法准确识别的问题，降低用户体验。
上述内容仅用于辅助理解本发明的技术方案，并不代表承认上述内容是现有技术。
发明内容
本发明的主要目的在于提供一种语音控制方法及智能电视，旨在解决现有技术中智能电视仅通过语音识别的方式进行控制带来的容易识别错误，且无法做到利用自然语义理解语音内容造成用户体验差的问题。
为实现上述目的，本发明提供的一种语音控制方法，包括：
智能电视接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
所述智能电视根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
所述智能电视根据所述语义信息对所述智能电视进行控制；
其中，所述智能电视根据预先设置的标准编码对照表对所述文本信息进行语义分析，包括：
根据所述文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定所述每一个文本字符所属的语言；
按照所述每一个文本字符所属的语言预先设置的标准编码对照表确定所述每一个文本字符的语义。
其中，所述智能电视根据所述语义信息对所述智能电视进行控制，包括：
判断当前已保存的语义数据库中是否包含所述语义信息；
当所述语义数据库中包含所述语义信息时，从所述语义数据库中获取与所述语义信息对应的操作数据，按照所述操作数据对所述智能电视进行控制，所述语义数据库中包含语义信息与操作数据之间的对应关系。
为实现上述目的，本发明提供的一种语音控制方法，包括：
智能电视接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
所述智能电视根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
所述智能电视根据所述语义信息对所述智能电视进行控制。
为了解决上述技术问题，本发明提供一种智能电视，包括：
接收模块，用于接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
分析模块，用于根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
控制模块，用于根据所述语义信息对所述智能电视进行控制。
本发明提供的语音控制方法中，智能电视在获取用户输入的语音信号的文本信息之后，根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，以得到文本信息的语义信息，并根据该语义信息对智能电视进行控制，通过对语音信号的文本信息进行语义分析，使得能够有效识别用户输入的语音信号的语义，实现对语音信号的自然语义的理解，增加识别准确率，使得能够与用户更好地进行互动，改善用户体验。
附图说明
图1为本发明第一实施例中语音控制方法的流程示意图；
图2为图1中智能电视根据预先设置的标准编码对照表对文本信息进行语义分析的步骤的细化流程示意图；
图3为本发明第二实施例的语音控制方法的流程示意图；
图4为本发明第三实施例的智能电视的功能模块示意图；
图5为图4中第三实施例中智能电视的分析模块402的细化功能模块示意图；
图6为本发明第四实施例的智能电视的功能模块示意图。
本发明目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
本发明提供的语音控制方法中，智能电视在获取用户输入的语音信号的文本信息之后，根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，以得到文本信息的语义信息，并根据该语义信息对智能电视进行控制，使得智能电视能够实现对语音信号的自然语义的理解，增加识别准确率，使得能够与用户更好地进行互动，改善用户体验。
为了更好的理解本发明实施例中的技术方案，下面将介绍本发明实施例中语音控制***，该语音控制***包括语音输入设备及智能电视。
在本发明实施例中，语音输入设备包含语音录入功能及语音识别功能，其中，语音录入功能能够由设置在语音输入设备上的麦克风实现，语音输入设备能够利用该麦克风采集用户的语音，并生成语音信号。
其中，语音输入设备的语音识别功能可以由语音输入设备中的语音识别模块实现，麦克风在生成语音信号之后，将该语音信号传输给语音识别模块，语音识别模块对该语音信号进行识别，得到该语音信号的文本信息。
其中，语音输入设备可以是遥控器或者其他的移动终端，该其他的移动终端可以是智能手机、平板电脑、掌上电脑等智能终端设备。
其中，上述的语音识别模块识别用户输入的语音信号的方法可以是：将该语音信号进行PCM编码得到PCM码流，将得到的PCM码流与语音输入设备预先设置的PCM码流、文本字符及文本字符的字符编码之间的对应关系进行比较，确定每一个PCM码流对应的文本字符及文本字符的字符编码，以得到语音信号的文本信息，该文本信息中包含语音信号中的文本字符及文本字符的字符编码。例如：若用户输入的语音为“刘德华新闻”，则语音信号的PCM码流为：0x0100（刘）、0x0200（德）、0x0300（华）、0x0400（新）、0x0500（闻），括号中即为码流对应的文本字符。
在本发明实施例中，智能电视是搭载了操作***，可以自由安装和卸载软件程序，具有视频、娱乐、游戏等功能的终端，并可以通过有线或者无线的方式实现网络功能。
其中，语音输入设备与智能电视都具有无限通信模块，语音输入设备和智能电视通过各自的无线通信模块进行无线通信链接，从而建立智能电视与语音输入设备之间的无线语音通道。其中，无线通信模块可以是WIFI模块、蓝牙模块、或无限USB模块等等，此处不做限定。
请参阅图1，为本发明第一实施例中语音控制方法，该方法包括：
步骤101、智能电视接收语音输入设备发送的文本信息，文本信息为语音输入设备对用户输入的语音信号进行语音识别得到的；
在本发明实施例中，语音输入设备在录入用户输入的语音信号并对该语音信号进行语音识别得到该语音信号的文本信息之后，将该语音信号的文本信息发送给智能电视，其中，智能电视与语音输入设备可以通过上述的无线语音通道进行语音信号的文本信息的传输。
其中，智能电视将接收到该语音信号的文本信息，该文本信息中包含语音信号中文本字符及文本字符的字符编码。
步骤102、智能电视根据预先设置的语言的标准编码对照表对文本信息进行语义分析，得到文本信息的语义信息；
在本发明实施例中，智能电视在接收到语音信号的文本信息之后，将根据预先设置的语言的标准编码对照表对文本信息进行语义分析，得到文本信息的语义信息。
需要说明的是，智能电视中可以预先设置多种不同的语言的标准编码对照表，例如，对于中文可以预先设置统一码（Unicode）码对照表，对于英文可以预先设置美国标准信息交换代码（ASCII, American Standard Code for Information Interchange）码对照表。
其中，Unicode是一种在计算机上使用的字符编码，它为每种语言中的每个字符设定了统一并且唯一的二进制编码，且优选地，对于中文使用的标准编码对照表具体可以使用UTF-8（8-bit Unicode Transformation Format），UTF-8是针对Unicode的可变长度字符编码，是Unicode的实现方式之一。
其中，ASCII是基于拉丁字母的一套电脑编码***，主要用于显示现代英语和其他西欧语言。
步骤103、智能电视根据语义信息对智能电视进行控制。
在本发明实施例中，智能电视在解析得到语义信息之后，可根据该语义信息对智能电视进行控制。
本发明提供的语音控制方法中，智能电视在获取用户输入的语音信号的文本信息之后，根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，以得到文本信息的语义信息，并根据该语义信息对智能电视进行控制，使得能够有效识别用户输入的语音信号的语义，实现对语音信号的自然语义的理解，增加识别准确率，使得能够与用户更好地进行互动，改善用户体验。
请参阅图2，为本发明第一实施例中的步骤102智能电视根据预先设置的标准编码对照表岁文本信息进行语义分析的细化步骤的流程图，包括：
步骤201、根据文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定每一个文本字符所属的语言；
步骤202、按照每一个文本字符所属的语言预先设置的标准编码对照表确定每一个文本字符的语义。
在本发明实施例中，智能电视能够根据文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定每一个文本字符所属的语言，且还可按照每一个文本字符所属的语言预先设置的标准编码对照表确定每一个文本字符的语义。
需要说明的是，对于每一种语言的标准编码对照表，其字符的大小范围是不一样的，且是分区域的。
为了更好的理解本发明实施例中的语义分析过程。下面将以智能电视预先设置的语言为中文和英文为例进行介绍。
对于步骤201中的根据文本信息中的每一个文本字符的字符编码大小确定每一个文本字符所属的语言，具体可以为：智能电视依次判断文本信息中的文本字符的字符编码是否大于255，当某一个文本字符的字符编码大于255时，确定该文本字符属于中文字符，当某一个文本字符的字符编码小于255时，确定该文本字符属于英文字符。因为Unicode编码规则中汉字的字符大小均大于255，ASCII中的英文字符的大小均小于255，其中，英文字符包括英文单词、字母及***数字。
对于步骤202中的按照每一个文本字符所属的语言的标准编码对照表确定每一个文本字符的语义，具体可以为：智能电视依次遍历文本信息中的文本字符，当遍历到的文本字符为中文字符时，使用预先设置的Unicode编码对照表对该文本字符进行中文差分筛选模糊匹配计算，确定该文本字符的语义，当遍历到的文本字符属于英文字符时，使用预先设置的ASCII编码对照表对该文本字符进行英文差分筛选模糊匹配计算，确定该文本字符的含义。通过上述的遍历及匹配的过程，能够确定文本信息的语义信息。
其中，使用预先设置的Unicode编码对照表对字符进行中文差分筛选模糊匹配计算，确定该字符的语义，具体可以是：当待匹配文本字符为中文时，该文本字符的字符编码为Unicode的可变长度字符编码，依次计算该待匹配文本字符的可变长度字符编码与预先设置的Unicode编码对照表中的所有中文字符的可变长度字符编码之间的差值，当确定预先设置的Unicode编码对照表中的第N个中文字符的可变长度字符编码与该待匹配文本字符的可变长度字符编码之间的差值为0，则确定该待匹配文本字符即为该第N个中文字符，将该第N个中文字符的语义作为该待匹配文本字符的语义。且停止对该待匹配文本字符的中文差分筛选模糊匹配计算。例如：若待匹配文本字符的可变长度字符编码为0x5218，则将该可变长度字符编码依次与Unicode编码对照表中的中文字符的可变字符编码相减，当确定0x5218与第100个中文字符的可变长度字符编码0x5218的差值为0时，确定该第100个中文字符对应的语义（该语义为：刘）即为该待匹配文本字符的语义。
其中，使用预先设置的ASCII编码对照表对该字符进行英文差分筛选模糊匹配计算，确定该字符的含义，具体可以为：若待匹配文本字符为英文字符，则该待匹配文本字符的字符编码为ASCII码，依次计算该待匹配文本字符的ASCII码与预先设置的ASCII编码对照表中的所有英文字符的ASCII码之间的差值，当确定预先设置的ASCII编码对照表中的第M个英文字符的ASCII码与该待匹配字符的ASCII之间的差值为0，则确定该待匹配字符即为该第N个中文字符，将该第N个中文字符的语义作为该待匹配文本字符的语义。且停止对该待匹配文本字符的英文差分筛选模糊匹配计算。例如：若待匹配文本字符的ASCII码为49，则将该ASCII码49依次与ASCII码中的英文字符的ASCII码相减，当确定ASCII码49与第50个英文字符的ASCII码49差值为0时，确定该第50个英文字符对应的语义（该语义为：1）即为该待匹配文本字符的语义。
需要说明的是，在本发明实施例中，是以先确定文本信息中的每个文本字符所属的语言，然后再依次对每个文本字符进行模糊匹配计算为例描述语义信息的识别过程的，在实际应用中，还可以直接遍历文本信息中的文本字符，先判断遍历到的文本字符的字符编码是否大于255，并根据判断结果进行模糊匹配计算，以确定文本信息的语义信息。
需要说明的是，若在本发明实施例中，是以智能电视预先设置的语言为中文和英文进行描述的，在实际应用中，若语音输入设备输入的语音信号使用的语言不是智能电视当前使用的语言，智能电视可以先将语音识别后的语音信号的文本信息转换成按照智能电视预先设置的语言能够识别的文本信息。例如：若用户是利用摩纳哥语进行语音输入，则语音输入设备在进行语音识别之后，可以将识别到的摩纳哥语的文本信息转换为中文和/或英文文本信息。
可以理解的是：由于在不同的环境及不同的国家使用的智能电视中，可以预先设置多种语言的标准编码对照表，则用户可通过选择确定智能电视使用的语言之后，智能电视可以以其使用的语言的标准编码对照表执行上述步骤201和步骤202，而非预先设置的语言的标准编码对照表。
在本发明实施例中，通过在智能电视中预先设置不同语言使用的标准编码对照表，使得智能电视能够基于不同语言的标准编码对照表确定语音信号的文本信息中的每一个文本字符的语义，有效完成对语义的识别。且语义识别的准确率高，能够给用户更好的体验。
请参阅图3，为本发明第二实施例中的语音控制方法，包括：
步骤301、智能电视接收语音输入设备发送的文本信息，文本信息为语音输入设备对用户输入的语音信号进行语音识别得到的；
在本发明实施例中，步骤301中描述的内容与图1所示第一实施例中的步骤101描述的内容相似，此处不做赘述。
步骤302、智能电视根据预先设置的语言的标准编码对照表对文本信息进行语义分析，以得到文本信息的语义信息；
在本发明实施例中，步骤302描述的内容与图1所示第一实施例中步骤102描述的内容相似，且图2所示的步骤101的细化流程也是本发明中的步骤302的细化流程，此处不做赘述。
步骤303、判断当前已保存的语义数据库中是否包含语义信息；当语义数据库中包含语义信息时，执行步骤304、当语义信息库中未包含语义信息时，则执行步骤305；
在本发明实施例中，智能电视在通过语义分析得到文本信息的语义信息之后，将判断在已保存的语义数据库中是否包含该语义信息。
需要说明的是，该语义数据库中包含了语义信息与操作数据之间的对应关系，例如：若语义信息是调大音量，则与该调大音量对应的操作数据则为生成音量调大指令。若语音信息为周杰伦电影，则与该周杰伦电影对应的操作数据为将与周杰伦电影对应的网络链接地址的图片和/或文字信息显示在智能电视的显示界面上。因此，在语义数据库中不同的语义信息对应的操作数据也是不同的。
步骤304、从语义数据库中获取与语义信息对应的操作数据，按照操作数据对智能电视进行控制，语义数据库中包含语义信息与操作数据之间的对应关系;
在本发明实施例中，当语义数据库中包含语义信息时，智能电视将从该语义数据库中获取与该语义信息对应的操作数据，并按照该操作数据对智能电视进行控制，例如：若语义信息是降低音量，则与该降低音量对应的操作数据则为生成音量降低指令，智能电视将根据该音量降低指令降低预先设置大小的音量。若语音信息为刘德华电影，则与该刘德华电影对应的操作数据为将与刘德华电影对应的网络链接地址的图片和/或文字信息显示在智能电视的显示界面上，智能电视将从语音数据库中读取该刘德华电影对应的网络链接地址，并将该网络链接地址的图片和/或文字信息显示在智能电视的显示界面上，使得用户可以选择具体的电影进行观看。
步骤305、将语义信息按照语义进行分词处理，得到语义信息中包含的词语；
步骤306、将语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
步骤307、将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
步骤308、基于网络爬虫机制，根据关键词及关键词匹配的分类条目获取关键词在分类条目下对应的网络链接地址，网络链接地址为关键词在分类条目下的点击次数满足预先设置的点击次数的地址；
在本发明实施例中，当语义数据库中未包含该语义信息时，智能电视将语义信息按照语义进行分词处理，得到语义信息中包含的词语，例如：对于语义信息“刘德华新闻”，智能电视按照语义划分可以得到词语“刘德华”及“新闻”。
且在得到语义信息包含的词语之后，智能电视将该语义信息中包含的词语与预先设置的媒体数据进行匹配，其中，该媒体数据包含不同的分类条目，且每一个分类条目都对应着一个词语集合，例如：媒体数据包含的分类条目为：新闻、电影、电视剧、综艺、音乐等等分类条目，在综艺分类条目下，其词语集合中包含“刘德华、周杰伦、爸爸去哪儿……”等词语。
需要说明的是，在一个分类条目下，一个词语仅仅出现一次。
对于语义信息中的每一个词语，智能电视将该词语与媒体数据中的每一个分类条目对应的词语集合中的词语进行匹配，以确定该词语匹配的分类条目，及匹配的分类条目的个数。例如，对于词语“刘德华”，匹配的分类条目包括：电影、明星、综艺、音乐，匹配的分类条目的个数为4个。
在本发明实施例中，智能电视在得到语义信息中的词语在媒体数据的不同分类条目中匹配的分类条目的个数之后，将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词，其中，匹配的分类条目的个数满足预先设置的数值条件的词语是指匹配非分类条目的个数最大的词语，或者匹配的分类条目的个数排在前N的词语，其中，N为正整数。例如，词语集合中包含的词语“刘德华”的匹配分类条目的个数为4个，词语“新闻”匹配的分类条目的个数为1个，则确定匹配的分类条目的个数最大的词语“刘德华”为关键词。
在本发明实施例中，智能电视在得到关键词之后，将基于网络爬虫机制，根据关键词及关键词匹配的分类条目获取该关键词在与其匹配的不同的分类条目下对应的网络链接地址，该网络链接地址为基于该关键词在与其匹配的不同分类条目下的点击次数满足预先设置的点击次数的地址。
其中，满足预先设置的点击次数的地址可以是点击次数排序排在前N的地址，该N为正整数。
其中，智能电视基于网络爬虫机制获取该关键词在与其匹配的不同分类条目下对应的网络链接地址，具体可以为：智能电视将关键词及关键词匹配的分类条目提供给网络爬虫，网络爬虫将基于该关键词及该关键词匹配的分类条目，从云端服务器获取该关键词在与其匹配的不同的分类条目下的网络连接地址，并统计在上述不同的分类条目下每一个网络链接地址的点击次数，并对于上述的每一个分类条目下的网络链接地址的点击次数进行排序，将分类条目下排序处于前N的N个网络链接地址作为与该关键词在该分类条目下对应的网络链接地址。例如：关键词为“刘德华”，与该关键词匹配的分类条目包括：电影、电视、音乐、综艺，网络爬虫将利用该关键词“刘德华”及其分类条目，抓取该关键词“刘德华”分别在电影、电视、音乐、综艺等分类条目下的网络链接地址，并对不同的分类条目的网络链接地址进行统计及排序，确定在电影分类条目下点击次数排在前4的4个网络链接地址为关键词“刘德华”在电影分类条目下对应的网络链接地址，且关键词“刘德华”在电视、音乐、综艺等分类条目下对应的网络链接地址也可以确定。
步骤309、将关键词在分类条目下对应的网络链接地址对应的图片和/或文字显示在智能电视的显示界面上。
在本发明实施例中，智能电视在得到与关键词在与其匹配的分类条目下对应的网络链接地址之后，将该关键词在与其匹配的分类条目对应的网路链接地址对应的图片和/或文字显示在智能电视的显示节目上，使得用户能够对显示的内容进行选择。
进一步的，智能电视还可执行步骤310，即：
步骤310、将网络链接地址作为语义信息的操作数据，并将语义信息与网络链接地址的对应关系保存在语义数据库中。
在本发明实施例中，智能电视在获取到语义信息中的关键词对应的网络链接地址之后，将该网络链接地址作为语义信息的操作数据，并将该语义信息与网络链接地址之间的对应关系保存在语义数据库中。
需要说明的是，智能电视可以在执行步骤308之后，同时执行步骤309和步骤310，或者可以在执行步骤308之后，继续执行步骤309，并在执行步骤309之后再执行步骤310，或者可以在执行步骤308之后，先执行步骤310再执行步骤309，即步骤309和步骤310的执行的先后顺序并不限定。
且为了能够为用户提供更好的体验，当语义数据库未清理时间达到预先设置的时长时，智能电视还将按照预先设置的策略定期清理语义数据库，且清理的周期可以预先设置，例如可以设置一个星期清理一次，也可以设置一个月清理一次，具体的清理周期并不限定。
优选地，为了更好的管理语义数据库，还可以将语义数据库中的语义信息进行分类，例如，可以分为针对智能电视的音量、频道、开关等的语义信息，及针对智能电视的显示的内容的语义信息。
优选地，智能电视可以按照预先设置的清理策略对语义数据库进行清理，具体的，可以是完全清空语义数据库，或者是可以针对语义数据库中的某一类语义信息及对应的操作数据进行清理，例如可以仅仅对智能电视的显示的内容的语义信息的清理。
优选的，为了便于确定用户的使用习惯，智能电视在每次判断解析到语义信息是否包含在语义数据库中之后，若确定该语义信息包含在语义数据库中，则将该语义信息的计数值加1，以确定用户使用语音控制***的习惯及爱好。
且在对语义数据库进行清理的时候，还可以根据语义数据库中的语义信息的计数值进行清理，例如可以将计数值小于某一个数值的语义信息及其对应的操作数据清理掉。
优选的，在对语义信息的计数值加1时，还可以记录对该语义信息的计数值加1的时间点，使得在对语义数据库进行清理时，可以通过语义信息的记录的最近时间点距离当前时间之间的时长，若该时长大于某预先设置的时长，则删除该语义信息及其对应的操作数据。若该时长小于或等于某预先设置的时长，则保留该语音信息及其对应的操作数据。
在本发明实施例中，可以通过预先设置语义数据库，使得在分析得到文本信息的语义信息之后，若在语义数据库中可查找到该语义信息，则从该语义数据库中获取与该语义信息对应的操作数据，若在语义数据库中不包含该语义信息，则从网络中抓取该语义信息中包含的关键词在与其匹配的分类条目下对应的网络链接地址，以获取该语义信息的操作数据。由于语义信息的分析识别度高，不易出错，使得对智能电视执行的操作都是用户想要的操作，能够有效实现智能电视对用户自然语义的理解，改善用户体验。
请参阅图4，为本发明第三实施例中智能电视的功能模块示意图，包括：
接收模块401，用于接收语音输入设备发送的文本信息，文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
分析模块402，用于根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，得到该文本信息的语义信息；
控制模块403，用于根据该语义信息对该智能电视进行控制。
在本发明实施例中，智能电视在获取用户输入的语音信号的文本信息之后，根据预先设置的语言的标准编码对照表对该文本信息进行语义分析，以得到文本信息的语音信息，并根据该语义信息对智能电视进行控制，使得能够有效识别用户输入的语音信号的语义，实现对语音信号的自然语义的理解，增加识别准确率，使得能够与用户更好地进行互动，改善用户体验。
请参阅图5，为本发明图4所示第三实施例中的分析模块402的细化功能模块，包括：
第一确定模块501，用于根据所述文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定所述每一个文本字符所属的语言；
语义分析模块502，用于按照该每一个文本字符所属的语言预先设置的标准编码对照表确定该每一个文本字符的语义。
其中，第一确定模块501具体用于：
当预先设置的语言为中文和英文时，依次判断文本信息中的文本字符是否大于255；
若该文本字符大于255，则确定该文本字符属于中文字符；
若该文本字符小于255，则确定该文本字符属于英文字符。
在具体实现过程中，分析模块中的各个细化功能模块与图2中各个方法步骤中的操作相对应。有关这些操作步骤的具体内容已经在前文做了详细的描述，因此此处不再赘述。
在本发明实施例中，通过在智能电视中预先设置不同语言使用的标准编码对照表，使得智能电视能够基于该不同语言的标准编码对照表确定语音信号的文本信息中的每一个文本字符的语义，有效完成对语义的识别。且语义识别的准确率高，能够给用户更好的体验。
请参阅图6，为本发明第四实施例的智能电视的功能模块的示意图，包括：如图4所示第三实施例中的接收模块401、分析模块402，及控制模块403，且与图4所示实施例中描述的内容相似，此处不做赘述。
其中，控制模块403包括：
判断模块601，用于判断当前已保存的语义数据库中是否包含语义信息；
第一处理模块602，用于当该语义数据库中包含该语义信息时，从该语义数据库中获取与该语义信息对应的操作数据，按照该操作数据对该智能电视进行控制，该语义数据库中包含语义信息与操作数据之间的对应关系。
其中，控制模块403还包括：
分词模块603，用于当语义数据库中未包含语义信息时，将语义信息按照语义进行分词处理，得到语义信息中包含的词语；
匹配模块604，用于将语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
关键词确定模块605，用于将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
地址获取模块606，用于基于网络爬虫机制，根据关键词及关键词匹配的分类条目获取关键词在分类条目下对应的网络链接地址，网络链接地址为关键词在分类条目下的点击次数满足预先设置的点击次数的地址；
显示模块607，用于将关键词在分类条目下对应的网络链接地址对应的图片和/或文字显示在智能电视的显示界面上。
在本发明实施例中，智能电视还包括：
保存模块608，用于将网络链接地址作为语义信息的操作数据，并将该语义信息与该网络链接地址的对应关系保存在语义数据库中。
清理模块609，用于当该语义数据库未清理时间达到预先设置的时长时，按照预先设置的清理策略清理语义数据库。
在具体实现过程中，本实施例中的智能电视中的各个模块与图3中各个方法步骤中的操作相对应。有关这些操作步骤的具体内容已经在前文做了详细的描述，因此此处不再赘述。
图6所示的第四实施例中，清理模块609是与保存模块608连接，表示清理模块609的功能可以在保存模块608的功能之后执行，实际上，需要说明的是，清理模块609的功能还可以在图6所示的第四实施例中除保存模块608之外的其他任何模块的功能执行之前或者之后执行，即不限定清理模块609与其他模块之间的连接关系，只要语义数据库未清理的时间达到预先设置的时长，即可启动清理模块609按照预先设置的清理策略清理语义数据库。
在本发明实施例中，可以通过预先设置语义数据库，使得在分析得到文本信息的语义信息之后，若在语义数据库中可查找到该语义信息，则从该语义数据库中获取与该语义信息对应的操作数据，若在语义数据库中不包含该语义信息，则从网络中抓取该语义信息中包含的关键词在与其匹配的分类条目下对应的网络链接地址，以获取该语义信息的操作数据。由于语义信息的分析识别度高，不易出错，使得对智能电视执行的操作都是用户想要的操作，能够有效实现智能电视对用户自然语义的理解，改善用户体验。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质（如ROM/RAM、磁碟、光盘）中，包括若干指令用以使得一台终端设备（可以是手机，计算机，服务器，空调器，或者网络设备等）执行本发明各个实施例的方法。
以上仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。

Claims (16)

  1. 一种语音控制方法，其特征在于，包括：
    智能电视接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
    所述智能电视根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
    所述智能电视根据所述语义信息对所述智能电视进行控制；
    其中，
    所述智能电视根据预先设置的标准编码对照表对所述文本信息进行语义分析，包括：
    根据所述文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定所述每一个文本字符所属的语言；
    按照所述每一个文本字符所属的语言预先设置的标准编码对照表确定所述每一个文本字符的语义；
    其中，所述智能电视根据所述语义信息对所述智能电视进行控制，包括：
    判断当前已保存的语义数据库中是否包含所述语义信息；
    当所述语义数据库中包含所述语义信息时，从所述语义数据库中获取与所述语义信息对应的操作数据，按照所述操作数据对所述智能电视进行控制，所述语义数据库中包含语义信息与操作数据之间的对应关系。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    当所述语义数据库中未包含所述语义信息时，将所述语义信息按照语义进行分词处理，得到所述语义信息中包含的词语；
    将所述语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定所述语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
    将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
    基于网络爬虫机制，根据所述关键词及所述关键词匹配的分类条目获取所述关键词在所述分类条目下对应的网络链接地址，所述网络链接地址为所述关键词在所述分类条目下的点击次数满足预先设置的点击次数的地址；
    将所述关键词在所述分类条目下对应的网络链接地址对应的图片和/或文字显示在所述智能电视的显示界面上。
  3. 根据权利要求2所述的方法，其特征在于，所述方法还包括：
    将所述网络链接地址作为所述语义信息的操作数据，并将所述语义信息与所述网络链接地址的对应关系保存在所述语义数据库中；
    当所述语义数据库未清理时间达到预先设置的时长时，按照预先设置的清理策略清理所述语义数据库。
  4. 一种语音控制方法，其特征在于，包括：
    智能电视接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
    所述智能电视根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
    所述智能电视根据所述语义信息对所述智能电视进行控制。
  5. 根据权利要求4所述的方法，其特征在于，所述智能电视根据预先设置的标准编码对照表对所述文本信息进行语义分析，包括：
    根据所述文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定所述每一个文本字符所属的语言；
    按照所述每一个文本字符所属的语言预先设置的标准编码对照表确定所述每一个文本字符的语义。
  6. 根据权利要求4所述的方法，其特征在于，所述智能电视根据所述语义信息对所述智能电视进行控制，包括：
    判断当前已保存的语义数据库中是否包含所述语义信息；
    当所述语义数据库中包含所述语义信息时，从所述语义数据库中获取与所述语义信息对应的操作数据，按照所述操作数据对所述智能电视进行控制，所述语义数据库中包含语义信息与操作数据之间的对应关系。
  7. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    当所述语义数据库中未包含所述语义信息时，将所述语义信息按照语义进行分词处理，得到所述语义信息中包含的词语；
    将所述语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定所述语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
    将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
    基于网络爬虫机制，根据所述关键词及所述关键词匹配的分类条目获取所述关键词在所述分类条目下对应的网络链接地址，所述网络链接地址为所述关键词在所述分类条目下的点击次数满足预先设置的点击次数的地址；
    将所述关键词在所述分类条目下对应的网络链接地址对应的图片和/或文字显示在所述智能电视的显示界面上。
  8. 根据权利要求7所述的方法，其特征在于，所述方法还包括：
    将所述网络链接地址作为所述语义信息的操作数据，并将所述语义信息与所述网络链接地址的对应关系保存在所述语义数据库中；
    当所述语义数据库未清理时间达到预先设置的时长时，按照预先设置的清理策略清理所述语义数据库。
  9. 一种智能电视，其特征在于，包括：
    接收模块，用于接收语音输入设备发送的文本信息，所述文本信息为所述语音输入设备对用户输入的语音信号进行语音识别得到的；
    分析模块，用于根据预先设置的语言的标准编码对照表对所述文本信息进行语义分析，得到所述文本信息的语义信息；
    控制模块，用于根据所述语义信息对所述智能电视进行控制。
  10. 根据权利要求9所述的智能电视，其特征在于，所述分析模块包括：
    第一确定模块，用于根据所述文本信息中的每一个文本字符的字符编码大小及预先设置的语言的字符大小区域确定所述每一个文本字符所属的语言；
    语义分析模块，用于按照所述每一个文本字符所属的语言预先设置的标准编码对照表确定所述每一个文本字符的语义。
  11. 根据权利要求9所述的智能电视，其特征在于，所述控制模块包括：
    判断模块，用于判断当前已保存的语义数据库中是否包含所述语义信息；
    第一处理模块，用于当所述语义数据库中包含所述语义信息时，从所述语义数据库中获取与所述语义信息对应的操作数据，按照所述操作数据对所述智能电视进行控制，所述语义数据库中包含语义信息与操作数据之间的对应关系。
  12. 根据权利要求11所述的智能电视，其特征在于，所述控制模块还包括：
    分词模块，用于当所述语义数据库中未包含所述语义信息时，将所述语义信息按照语义进行分词处理，得到所述语义信息中包含的词语；
    匹配模块，用于将所述语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定所述语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
    关键词确定模块，用于将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
    地址获取模块，用于基于网络爬虫机制，根据所述关键词及所述关键词匹配的分类条目获取所述关键词在所述分类条目下对应的网络链接地址，所述网络链接地址为所述关键词在所述分类条目下的点击次数满足预先设置的点击次数的地址；
    显示模块，用于将所述关键词在所述分类条目下对应的网络链接地址对应的图片和/或文字显示在所述智能电视的显示界面上。
  13. 根据权利要求12所述的智能电视，其特征在于，所述智能电视还包括：
    保存模块，用于将所述网络链接地址作为所述语义信息的操作数据，并将所述语义信息与所述网络链接地址的对应关系保存在所述语义数据库中；
    清理模块，用于当所述语义数据库未清理时间达到预先设置的时长时，按照预先设置的清理策略清理所述语义数据库。
  14. 根据权利要求10所述的智能电视，其特征在于，所述控制模块包括：
    判断模块，用于判断当前已保存的语义数据库中是否包含所述语义信息；
    第一处理模块，用于当所述语义数据库中包含所述语义信息时，从所述语义数据库中获取与所述语义信息对应的操作数据，按照所述操作数据对所述智能电视进行控制，所述语义数据库中包含语义信息与操作数据之间的对应关系。
  15. 根据权利要求14所述的智能电视，其特征在于，所述控制模块还包括：
    分词模块，用于当所述语义数据库中未包含所述语义信息时，将所述语义信息按照语义进行分词处理，得到所述语义信息中包含的词语；
    匹配模块，用于将所述语义信息中包含的词语依次与预先设置的媒体数据中不同的分类条目对应的词语集合中的词语进行匹配，确定所述语义信息中的每一个词语匹配的分类条目，及匹配的分类条目的个数；
    关键词确定模块，用于将匹配的分类条目的个数满足预先设置的数值条件的词语作为关键词；
    地址获取模块，用于基于网络爬虫机制，根据所述关键词及所述关键词匹配的分类条目获取所述关键词在所述分类条目下对应的网络链接地址，所述网络链接地址为所述关键词在所述分类条目下的点击次数满足预先设置的点击次数的地址；
    显示模块，用于将所述关键词在所述分类条目下对应的网络链接地址对应的图片和/或文字显示在所述智能电视的显示界面上。
  16. 根据权利要求15所述的智能电视，其特征在于，所述智能电视还包括：
    保存模块，用于将所述网络链接地址作为所述语义信息的操作数据，并将所述语义信息与所述网络链接地址的对应关系保存在所述语义数据库中；
    清理模块，用于当所述语义数据库未清理时间达到预先设置的时长时，按照预先设置的清理策略清理所述语义数据库。
PCT/CN2016/085157 2016-03-18 2016-06-07 语音控制方法及智能电视 WO2017156893A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610158350.X 2016-03-18
CN201610158350.XA CN105764185B (zh) 2016-03-18 2016-03-18 交流驱动混合调光电路和电视机

Publications (1)

Publication Number Publication Date
WO2017156893A1 true WO2017156893A1 (zh) 2017-09-21

Family

ID=56345323

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/CN2016/084866 WO2017156891A1 (zh) 2016-03-18 2016-06-04 交流驱动混合调光电路和电视机
PCT/CN2016/085157 WO2017156893A1 (zh) 2016-03-18 2016-06-07 语音控制方法及智能电视

Family Applications Before (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/084866 WO2017156891A1 (zh) 2016-03-18 2016-06-04 交流驱动混合调光电路和电视机

Country Status (2)

Country Link
CN (1) CN105764185B (zh)
WO (2) WO2017156891A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107742519A (zh) * 2017-10-31 2018-02-27 珠海市美瑞华医用科技有限公司 一种智能控制输液设备的语音输入***
CN109388634A (zh) * 2018-09-18 2019-02-26 平安科技（深圳）有限公司 地址信息的处理方法、终端设备及计算机可读存储介质
CN111508494A (zh) * 2020-04-20 2020-08-07 广东工业大学 一种智能纳税语音咨询方法和***
CN112256947A (zh) * 2019-07-05 2021-01-22 北京猎户星空科技有限公司 一种推荐信息的确定方法、装置、***、设备及介质
CN112422377A (zh) * 2020-10-26 2021-02-26 深圳Tcl新技术有限公司 智能终端的控制方法、装置及计算机可读存储介质
CN113488049A (zh) * 2021-07-07 2021-10-08 杭州贯农科技有限公司 一种插件式语音识别方法、智能电子秤及交易平台
CN113658598A (zh) * 2021-08-12 2021-11-16 海信电子科技（深圳）有限公司 一种显示设备的语音交互方法及显示设备
CN113778226A (zh) * 2021-08-26 2021-12-10 江西恒必达实业有限公司 一种基于语音识别技术控制智能家居的红外ai智能眼镜
WO2023246151A1 (zh) * 2022-06-22 2023-12-28 海信视像科技股份有限公司 显示设备和控制方法

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106455209B (zh) * 2016-09-30 2018-12-07 深圳Tcl数字技术有限公司 调光接口电路及交流驱动电路
CN106793328A (zh) * 2017-01-06 2017-05-31 大连工业大学 基于峰值电流可调的选择调光装置及其方法
CN107750076B (zh) * 2017-11-22 2024-04-23 深圳中电数码显示有限公司 一种led驱动电路
CN108235526B (zh) * 2018-04-03 2024-02-23 深圳茂硕电子科技有限公司 一种控制电路
CN110139445A (zh) * 2019-05-31 2019-08-16 深圳市奋勇光电有限公司 一种电量和调光控制二合一电路
CN111539408A (zh) * 2020-04-08 2020-08-14 王鹏 基于拍照识物的智能点读方案
WO2022226811A1 (zh) * 2021-04-27 2022-11-03 华为技术有限公司 构建语音识别模型和语音处理的方法和***
CN115438212B (zh) * 2022-08-22 2023-03-31 蒋耘晨 一种影像投射***、方法及设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080228480A1 (en) * 2007-03-13 2008-09-18 Shuhei Maegawa Speech recognition method, speech recognition system, and server thereof
CN102629246A (zh) * 2012-02-10 2012-08-08 北京百纳信息技术有限公司 识别浏览器语音命令的服务器及浏览器语音命令识别方法
CN102855874A (zh) * 2012-09-07 2013-01-02 深圳市信利康电子有限公司 基于互联网语音交互的家电控制方法及***
CN103219007A (zh) * 2013-03-27 2013-07-24 谢东来 语音识别方法及装置
CN104717536A (zh) * 2013-12-11 2015-06-17 中国电信股份有限公司 一种语音控制的方法和***
CN105161106A (zh) * 2015-08-20 2015-12-16 深圳Tcl数字技术有限公司 智能终端的语音控制方法、装置及电视机***
CN105512182A (zh) * 2015-11-25 2016-04-20 深圳Tcl数字技术有限公司 语音控制方法及智能电视

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100334754B1 (ko) * 1999-12-24 2002-05-04 이형도 냉음극 형광등용 인버터
US8026676B2 (en) * 2008-10-08 2011-09-27 Richtek Technology Corporation, R.O.C. Dimming control circuit
CN201349354Y (zh) * 2009-01-19 2009-11-18 冠捷投资有限公司 结合模拟及数字调光的背光源驱动装置
CN101605413B (zh) * 2009-07-06 2012-07-04 英飞特电子（杭州）有限公司 适用于可控硅调光的led驱动电路
US8911111B2 (en) * 2012-06-05 2014-12-16 Shenzhen China Star Optoelectronics Technology, Co., Ltd. LED backlight system and display device
CN203134330U (zh) * 2012-12-28 2013-08-14 深圳Tcl新技术有限公司 Led显示屏背光亮度调节电路及led电视
CN103474030B (zh) * 2013-09-11 2016-10-05 深圳市瀚强科技股份有限公司 一种隔离led背光驱动电路及led显示设备
CN104243872B (zh) * 2014-09-17 2018-04-24 广州视琨电子科技有限公司 低功耗的恒流及背光控制电路和电视机
CN204291476U (zh) * 2014-11-28 2015-04-22 深圳桑达国际电源科技有限公司 Led调光电源电路及led灯具
CN105163416B (zh) * 2015-05-02 2018-08-21 广东瑞德智能科技股份有限公司 一种改进的led灯用可控硅调光驱动电路

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080228480A1 (en) * 2007-03-13 2008-09-18 Shuhei Maegawa Speech recognition method, speech recognition system, and server thereof
CN102629246A (zh) * 2012-02-10 2012-08-08 北京百纳信息技术有限公司 识别浏览器语音命令的服务器及浏览器语音命令识别方法
CN102855874A (zh) * 2012-09-07 2013-01-02 深圳市信利康电子有限公司 基于互联网语音交互的家电控制方法及***
CN103219007A (zh) * 2013-03-27 2013-07-24 谢东来 语音识别方法及装置
CN104717536A (zh) * 2013-12-11 2015-06-17 中国电信股份有限公司 一种语音控制的方法和***
CN105161106A (zh) * 2015-08-20 2015-12-16 深圳Tcl数字技术有限公司 智能终端的语音控制方法、装置及电视机***
CN105512182A (zh) * 2015-11-25 2016-04-20 深圳Tcl数字技术有限公司 语音控制方法及智能电视

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107742519A (zh) * 2017-10-31 2018-02-27 珠海市美瑞华医用科技有限公司 一种智能控制输液设备的语音输入***
CN109388634A (zh) * 2018-09-18 2019-02-26 平安科技（深圳）有限公司 地址信息的处理方法、终端设备及计算机可读存储介质
CN109388634B (zh) * 2018-09-18 2024-05-03 平安科技（深圳）有限公司 地址信息的处理方法、终端设备及计算机可读存储介质
CN112256947B (zh) * 2019-07-05 2024-01-26 北京猎户星空科技有限公司 一种推荐信息的确定方法、装置、***、设备及介质
CN112256947A (zh) * 2019-07-05 2021-01-22 北京猎户星空科技有限公司 一种推荐信息的确定方法、装置、***、设备及介质
CN111508494A (zh) * 2020-04-20 2020-08-07 广东工业大学 一种智能纳税语音咨询方法和***
CN111508494B (zh) * 2020-04-20 2023-03-07 广东工业大学 一种智能纳税语音咨询方法和***
CN112422377A (zh) * 2020-10-26 2021-02-26 深圳Tcl新技术有限公司 智能终端的控制方法、装置及计算机可读存储介质
CN113488049A (zh) * 2021-07-07 2021-10-08 杭州贯农科技有限公司 一种插件式语音识别方法、智能电子秤及交易平台
CN113658598A (zh) * 2021-08-12 2021-11-16 海信电子科技（深圳）有限公司 一种显示设备的语音交互方法及显示设备
CN113658598B (zh) * 2021-08-12 2024-02-27 Vidaa(荷兰)国际控股有限公司 一种显示设备的语音交互方法及显示设备
CN113778226A (zh) * 2021-08-26 2021-12-10 江西恒必达实业有限公司 一种基于语音识别技术控制智能家居的红外ai智能眼镜
WO2023246151A1 (zh) * 2022-06-22 2023-12-28 海信视像科技股份有限公司 显示设备和控制方法

Also Published As

Publication number Publication date
WO2017156891A1 (zh) 2017-09-21
CN105764185B (zh) 2017-12-12
CN105764185A (zh) 2016-07-13

Similar Documents

Publication Publication Date Title
WO2017156893A1 (zh) 语音控制方法及智能电视
WO2013139239A1 (en) Method for recommending users in social network and the system thereof
WO2014003283A1 (en) Display apparatus, method for controlling display apparatus, and interactive system
WO2017143692A1 (zh) 智能电视及其语音控制方法
WO2014107101A1 (en) Display apparatus and method for controlling the same
WO2014107097A1 (en) Display apparatus and method for controlling the display apparatus
WO2020251283A1 (en) Selecting artificial intelligence model based on input data
WO2020009297A1 (ko) 도메인 추출기반의 언어 이해 성능 향상장치및 성능 향상방법
WO2014107102A1 (en) Display apparatus and method of controlling display apparatus
WO2017111252A1 (en) Electronic device and method of scanning channels in electronic device
WO2016137167A1 (en) Terminal
WO2019061613A1 (zh) 贷款资质筛选方法、装置及计算机可读存储介质
WO2018006489A1 (zh) 终端的语音交互方法及装置
WO2017054592A1 (zh) 一种界面显示的方法及终端
WO2019177182A1 (ko) 속성 정보 분석을 통한 멀티미디어 컨텐츠 검색장치 및 검색방법
WO2018034426A1 (ko) 커널 ｒｄｒ을 이용한 태깅 말뭉치 오류 자동수정방법
WO2013170662A1 (zh) 添加好友信息的方法及装置、计算机存储介质
WO2015194693A1 (ko) 영상 표시 기기 및 그의 동작 방법
WO2016032021A1 (ko) 음성 명령 인식을 위한 장치 및 방법
WO2015129989A1 (ko) 콘텐츠 및 음원 추천 장치 및 방법
WO2019062112A1 (zh) 空调器控制方法、装置、空调器及计算机可读存储介质
WO2017054488A1 (zh) 电视播放控制方法、服务器及电视播放控制***
WO2019041851A1 (zh) 家电售后咨询方法、电子设备和计算机可读存储介质
WO2019085543A1 (zh) 电视机***及电视机控制方法
WO2020032564A1 (en) Electronic device and method for providing one or more items in response to user speech

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16894055

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 01.02.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16894055

Country of ref document: EP

Kind code of ref document: A1