WO2017028706A1 - 对比赛进行文字直播的方法及装置 - Google Patents

对比赛进行文字直播的方法及装置 Download PDF

Info

Publication number
WO2017028706A1
WO2017028706A1 PCT/CN2016/094115 CN2016094115W WO2017028706A1 WO 2017028706 A1 WO2017028706 A1 WO 2017028706A1 CN 2016094115 W CN2016094115 W CN 2016094115W WO 2017028706 A1 WO2017028706 A1 WO 2017028706A1
Authority
WO
WIPO (PCT)
Prior art keywords
game
historical
statement
live
action
Prior art date
Application number
PCT/CN2016/094115
Other languages
English (en)
French (fr)
Inventor
叶舟
王瑜
陈凡
杨洋
徐季秋
付志嵩
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Priority to US15/752,956 priority Critical patent/US10474749B2/en
Priority to JP2018507650A priority patent/JP2018525675A/ja
Priority to SG11201801334TA priority patent/SG11201801334TA/en
Priority to KR1020187007489A priority patent/KR20180041189A/ko
Publication of WO2017028706A1 publication Critical patent/WO2017028706A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/488—Data services, e.g. news ticker
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/488—Data services, e.g. news ticker
    • H04N21/4888—Data services, e.g. news ticker for displaying teletext characters
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3334—Selection or weighting of terms from queries, including natural language queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G06F16/337—Profile generation, learning or modification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/02—Arrangements for generating broadcast information; Arrangements for generating broadcast-related information with a direct linking to broadcast information or to broadcast space-time; Arrangements for simultaneous generation of broadcast information and broadcast-related information
    • H04H60/07—Arrangements for generating broadcast information; Arrangements for generating broadcast-related information with a direct linking to broadcast information or to broadcast space-time; Arrangements for simultaneous generation of broadcast information and broadcast-related information characterised by processes or methods for the generation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/235—Processing of additional data, e.g. scrambling of additional data or processing content descriptors
    • H04N21/2353—Processing of additional data, e.g. scrambling of additional data or processing content descriptors specifically adapted to content descriptors, e.g. coding, compressing or processing of metadata
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/439—Processing of audio elementary streams
    • H04N21/4394—Processing of audio elementary streams involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/488—Data services, e.g. news ticker
    • H04N21/4884—Data services, e.g. news ticker for displaying subtitles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81—Monomedia components thereof
    • H04N21/8126—Monomedia components thereof involving additional data, e.g. news, sports, stocks, weather forecasts
    • H04N21/8133—Monomedia components thereof involving additional data, e.g. news, sports, stocks, weather forecasts specifically related to the content, e.g. biography of the actors in a movie, detailed information about an article seen in a video program
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/025—Systems for the transmission of digital non-picture data, e.g. of text during the active part of a television frame
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/21—Server components or server architectures
    • H04N21/218—Source of audio or video content, e.g. local disk arrays
    • H04N21/2187—Live feed

Definitions

  • the present application relates to the field of Internet technologies, and in particular, to a method and apparatus for performing live broadcast of a game.
  • the machine-based text live method is only a direct statement of the game, for example, 2015-05-2810:15:19, Warriors - Rockets, Curry, three-point shooting, goals, the current live broadcast method is relatively dull Boring, the live broadcast of the game is almost no user to watch, the live broadcast quality is poor.
  • aspects of the present application provide a method and apparatus for text broadcast of a game, which is used to suppress the live broadcast of the game and improve the quality of the live broadcast of the game.
  • a method for texting a game including:
  • the live broadcast statement is output.
  • a method for processing a live game statement including:
  • the historical live speech is marked with a theme and an emotional score.
  • an apparatus for performing live broadcast of a game including:
  • a receiving module configured to receive game data in a current game in real time
  • a generating module configured to generate an emotionally-oriented real-time live speech describing the game data in the current game according to an emotionally-oriented historical live speech describing the game data in the historical game;
  • An output module configured to output the live broadcast statement.
  • an apparatus for processing a live game statement includes:
  • the obtaining module is configured to obtain a historical live speech statement with an emotional tendency used when the game data in the historical game is played live;
  • the mark scoring module is configured to perform topic mark and emotional scoring on the historical live speech according to the game action and the action result described by the historical live broadcast statement.
  • the game data in the current game is received in real time, and a live broadcast statement with emotional sentiment describing the game data in the current game is generated according to the historical live speech with emotional sentiment describing the game data in the historical game, and the live broadcast statement is generated.
  • Live live speech to achieve live broadcast of the current game. Since the application adopts a live speech with emotional inclination, it has an effect of suppressing the live broadcast of the current game, and improves the quality of the live broadcast of the game.
  • FIG. 1 is a schematic flowchart of a method for performing live text broadcast on a game according to an embodiment of the present application
  • FIG. 2 is a schematic flowchart of a method for processing a live game statement according to an embodiment of the present application
  • FIG. 3 is a schematic structural diagram of an apparatus for performing live text broadcast on a game according to an embodiment of the present application
  • FIG. 4 is a schematic structural diagram of an apparatus for performing live text broadcast on a game according to another embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of an apparatus for processing a live game statement according to an embodiment of the present disclosure.
  • the machine-based text live method is only a factual statement of the game, for example, 2015-05-28 10:15:19, Warriors-Rocket, Curry, three-point shooting, goals, live broadcast effect Emotionally relatively dull and boring, resulting in live broadcasts of almost no users to watch, the quality of live broadcast is poor.
  • the present application provides a method for text broadcast of a game, which belongs to a machine-based text live broadcast method, and the specific principle is: generating a description based on an emotionally-oriented historical live statement describing game data in a historical game.
  • a live-action live statement with emotional sentiment in the current game data to achieve live broadcast of the text of the current game. Since the present application uses a sentimental live speech in the live broadcast of the current game, the live broadcast of the current game has a swaying effect and improves the quality of the live broadcast of the game.
  • FIG. 1 is a schematic flowchart diagram of a method for performing live text broadcast on a game according to an embodiment of the present application. As shown in Figure 1, the method includes:
  • the game data in the current game is acquired in real time for real-time live broadcast of the current game.
  • the game data of the current game can be accessed from the competition party in real time, and can be provided by the official.
  • the game data mainly includes: information such as the game time, the game party, the game character, the game action, and the action result.
  • the game data in the Warriors' game against the Rockets are as follows:
  • the above game data is simply a narrative of the facts of the game, and there is no emotional fluctuation. If the game data is used for live broadcast of the game, the quality of the live broadcast is poor, and it will not be able to attract viewers to watch.
  • the present embodiment generates a real-time live speech with sentiment tendency, and uses the live broadcast statement with emotional tendency to perform live broadcast of the current game, so that the live broadcast effect has Emotional stagnation, attracting viewers to watch and improve the quality of live text, Improve the viewer's viewing experience and more.
  • the above-mentioned historical live broadcast statement may be a live broadcast sentence used when a live text broadcast of a historical game is performed by using a manual text live broadcast method, or may be a live broadcast sentence obtained by text-converting audio data generated by a live broadcast of a historical match by using an audio live broadcast method. Or it may be a live speech obtained by text conversion of video data generated by a live broadcast of a historical game by using a live video method.
  • the live speech statement used in the live broadcast of the text by the artificial text live broadcast method or the live speech generated by the audio and video live data is generally sentimental, and the live broadcast quality is good, as shown below:
  • the artificial text live broadcast method requires more human resources and higher implementation costs.
  • the text broadcast with emotional tendency can be realized, and the problem of human resource consumption and the like can be solved.
  • the live speech with the sentiment tendency used in the live broadcast of the text using the method of the embodiment may be obtained as the live broadcast. Statement.
  • the embodiment further provides a method for processing the live speech of the game. For pre-marking and emotional scoring of historical live speeches. As shown in Figure 2, the method includes:
  • the live broadcast sentence used when the text is broadcasted by the artificial typing method is used as the live broadcast statement used here.
  • a plurality of action themes are specifically set according to the game actions involved in the current game, and the basketball game is taken as an example, generally including “dribbling”, “passing the ball”, “going the basket”, “dunk”, and “rebounding”.
  • Action themes such as "caps”.
  • each action theme needs to be described by using some keywords, and according to these keywords, the corresponding live speech under each action theme can be determined.
  • the purpose of emotional scoring of historical live speech is to distinguish the emotional differences between different historical live speech sentences, and the historical live speech sentences with different emotional scores express different emotions.
  • the sentiment score of the historical live speech sentence may be determined by the action direction and degree described by the historical live speech statement, where the action direction actually refers to the action result, such as whether the shot is shot or not, or whether the dribble is smooth or not.
  • the action descriptor, the action direction descriptor, and the action degree descriptor may be parsed from the historical live speech; the game action described by the action descriptor, Marking the action theme of the historical live speech; determining the emotional score of the historical live speech according to the action direction described by the action direction descriptor and the action degree described by the action description word.
  • the action descriptor, action direction descriptor and action degree descriptor are parsed from the historical live speech: dribble, smooth, very.
  • the action description word the game action involved in the historical live speech is dribble, and the action theme of marking the historical live speech is: dribbling; further, the action direction and the action degree description word according to the action direction descriptor Describe the level of action and emotionally score the historical live speech.
  • the action direction described by the action direction descriptor and the action degree described by the action description word may be preset, and the strategy or rule for emotionally scoring the historical live speech is based on the set strategy or rule.
  • the live statement is used for emotional scoring.
  • the historical live speech after the topic mark and emotion score can be formed into a text live list and output for subsequent live broadcast of the game.
  • Ways include:
  • the historical live speech in which the emotion matches the game data in the current game is obtained from the historical live speech as the candidate historical live speech;
  • the candidate historical live broadcast statement is modified to obtain a live broadcast statement.
  • the set strategy or rule according to the game action and action result described in the game data of the current game and the action theme and emotional score of the historical live speech, the history of the emotion matching the game data in the current game is obtained from the historical live speech.
  • the live broadcast statement is used as a candidate historical live broadcast statement.
  • the pass-through action is applied to the live speech of the emotional scores 1-3 in the pass theme; the goal-like action is applied to the live speech of the emotional score of 3 to 5 in the shooting theme; , it is applicable to live speech statements with emotional scores of -3 to -5 in the shooting theme, and so on.
  • the game action and the action result described by the game data in the current game and the action theme and the emotional score of the historical live speech can be obtained, and the historical live broadcast in which the emotion matches the game data in the current game is obtained from the historical live speech.
  • the statement is used as a candidate historical live statement.
  • the candidate historical live broadcast statement may not be directly used as a live broadcast statement.
  • the candidate historical live broadcast sentence may be different from the match character in the current match's game data, so the candidate historical live broadcast statement may be Make corrections to replace the characters in the candidate historical live broadcast with the characters in the current game.
  • the candidate historical live speech can be adjusted, such as reducing an exclamation point, deleting a few words, etc., or adding a few words, etc., to obtain a real-time straightforward fit for the current game. Broadcast statement.
  • a candidate historical live broadcast sentence in which a plurality of emotions match the game data of the current game may be acquired, and the candidate historical live broadcast with the highest absolute value of the sentiment score may be selected from the plurality of candidate historical live broadcast sentences.
  • the candidate historical live speech with the smallest absolute value of the sentence or sentiment score or the candidate historical live speech with the text expression meeting the preset requirement is used as the target historical live broadcast statement, and the target historical live broadcast sentence is corrected to obtain a real-time live broadcast statement.
  • a candidate historical live speech with the highest absolute value of the sentiment score may be selected, for example, a candidate score of 5 from a candidate historical live sentence with a score of 3 to 5.
  • the candidate historical live statement is used as the target historical live statement.
  • a candidate historical live speech with the highest absolute value of the emotional score may be selected, for example, a candidate historical live sentence with a score of -3 to -5 is selected as a score.
  • -5 candidate historical live statement as the target historical live statement.
  • the candidate historical live speech with the lowest absolute value of the sentiment score may be selected, for example, the candidate historical live speech with a score of -2 to 2 is selected to have a score of 0.
  • the candidate historical live statement is used as the target historical live statement.
  • candidate historical live speech with the highest absolute value of the sentiment score or the candidate historical live speech with the smallest absolute value of the sentiment score or the candidate historical live speech with the text expression meeting the preset requirement are multiple, one can randomly select one as the target historical live broadcast. Statement.
  • the candidate historical live statement can be directly corrected to obtain a live broadcast statement.
  • the method provided by the present application can also comment on the game in real time in addition to the live broadcast of the current game by using the above-mentioned emotionally-oriented live speech.
  • a real-time comment statement with an emotional tendency to perform real-time comment on the current game is generated; and the real-time comment statement is output to perform real-time comment on the game. Further improve the quality of the live broadcast, increase the user's interest in the live broadcast, and attract viewers to watch the live broadcast.
  • a historical comment sentence whose emotion is consistent with the content of the game is obtained from the historical comment sentence as a candidate history comment statement;
  • the candidate history comment statement is modified to obtain a real-time comment statement.
  • the time point for commenting may be preset, and the game data broadcasted before the time point constitutes the content of the game to be commented; or may be preset to be commented.
  • Game scenarios such as each time a goal is scored, or each time a goal is lost, or when a team scores consecutive goals, or when a team loses the ball continuously, the match data related to the goal or goal may be As the content to be reviewed, etc.; or alternatively, a designated game character, such as Thompson, may be set, and when the game data related to the designated game character appears, the information related to the game character is used as the game content to be commented.
  • a policy or rule can also be set. According to the preset policy or rule, according to the action actions and action results involved in the content of the game and the action theme and the emotional score of the historical comment statement, a historical comment sentence whose emotion matches the content of the game is obtained from the historical comment statement as a candidate history. Comment statement.
  • the present embodiment broadcasts the current game through the live speech with emotional tendency, and the real-time comment on the current game through the emotional comment statement is beneficial to further improve the quality of the live text and attract more viewers to watch the live match. .
  • the method provided in this embodiment further includes: real-time statistics according to the game data of the current game. a list of information under each action topic involved in the game; receiving a game question sent by the viewer during the current game, obtaining an answer from the corresponding information list according to the action theme involved in the game, and outputting the game result at the interaction time of the current game.
  • real-time statistics according to the game data of the current game. a list of information under each action topic involved in the game
  • receiving a game question sent by the viewer during the current game obtaining an answer from the corresponding information list according to the action theme involved in the game, and outputting the game result at the interaction time of the current game.
  • the game data of the current game received in real time can be stored in the live database.
  • the live database there is a full amount of game data related to the current game, and different games have different types of data.
  • the specific implementation method is as follows: classify the game data in the live database according to the action theme involved in the current game. For example, for the theme of shooting, a list of information can be used to store the hit rate and the turnover rate of each player in the game. For the subject of passing, you can use a list of information to store information such as the number of passes per player in the game, in order to answer the questions asked by the audience.
  • the game question raised by the viewer can be analyzed, and the problem is determined to involve the shooting theme, so the corresponding result is searched in the background information list. For example, for the game question "How much is Curry hit so far?", the answer is "50%.”
  • the process of searching for answers can be realized by using a mature search engine architecture such as Lucene and Solr in technical implementation, and the storm flow computing framework can be applied in real-time computing, but is not limited thereto.
  • the interaction with the audience usually occurs when the game is suspended or during the intermission, which means that the interaction time of the current game refers to the game suspension or the intermission time. Since the audience interaction time is often limited, and the number of contest questions raised by the audience will be many, it is generally impossible to answer all the competition questions, so it is possible to adaptively select some of the competition questions to answer.
  • One of the most basic options is to choose the most popular game issues for display. For example, this Ayong Word2Vec model will cluster the game questions and answers, select the category with the most interactions, and extract one or several game questions to answer. Depending on the length of the interaction, you can answer different numbers of contest questions by heat.
  • the method for text broadcasting the game uses a sentimental live speech to broadcast the current game, and uses the emotionally inclined comment statement to comment on the current game in real time, and also supports interaction with the audience.
  • To answer the contest questions raised by the audience improve the quality of the live text, improve the audience's experience of watching live broadcasts, and basically get rid of the live broadcast, reduce the cost and improve the efficiency.
  • FIG. 3 is a schematic structural diagram of an apparatus for performing live text broadcast on a game according to an embodiment of the present application. As shown in FIG. 3, the apparatus includes: a receiving module 31, a generating module 32, and an output module 33.
  • the receiving module 31 is configured to receive game data in the current game in real time.
  • the generating module 32 is configured to generate a live broadcast statement with emotional sentiment describing the game data in the current game received by the receiving module 31 according to the historical live speech with emotional sentiment describing the game data in the historical game.
  • the output module 33 is configured to output a live live statement generated by the generating module 32.
  • the generating module 32 is specifically configured to:
  • the historical live speech in which the emotion matches the game data in the current game is obtained from the historical live speech as the candidate historical live speech;
  • the candidate historical live broadcast statement is modified to obtain a live broadcast statement.
  • the generating module 32 may specifically be used to:
  • the candidate historical live speech with the highest absolute value of the sentiment score or the candidate historical live speech with the lowest absolute value of the sentiment score or the text expression meets the preset requirement from the plurality of candidate historical live broadcast sentences.
  • the candidate historical live broadcast statement is used as the target historical live broadcast statement;
  • the target historical live broadcast statement is modified to obtain the live broadcast statement.
  • a candidate historical live speech with the highest absolute value of the sentiment score may be selected, for example, a candidate score of 5 from a candidate historical live sentence with a score of 3 to 5.
  • the candidate historical live statement is used as the target historical live statement.
  • a candidate historical live speech with the highest absolute value of the emotional score may be selected, for example, a candidate historical live sentence with a score of -3 to -5 is selected as a score.
  • -5 candidate historical live statement as the target historical live statement.
  • the candidate historical live speech with the lowest absolute value of the sentiment score may be selected, for example, the candidate historical live speech with a score of -2 to 2 is selected to have a score of 0.
  • the candidate historical live statement is used as the target historical live statement.
  • candidate historical live speech with the highest absolute value of the sentiment score or the candidate historical live speech with the smallest absolute value of the sentiment score or the candidate historical live speech with the text expression meeting the preset requirement are multiple, one can randomly select one as the target historical live broadcast. Statement.
  • the generating module 32 is further configured to generate an emotionally inclined real-time comment statement that reviews the current game in real time according to the emotionally inclined historical comment statement that reviews the historical game in real time.
  • the output module 33 is further configured to: output a real-time comment statement generated by the generating module 32.
  • the generating module 32 is configured to: when generating a real-time comment statement with an emotional tendency to perform real-time comment on the current game according to the historical comment statement with an emotional tendency to comment on the historical game in real time, specifically:
  • a historical comment sentence whose emotion is consistent with the content of the game is obtained from the historical comment sentence as a candidate history comment statement;
  • the candidate history comment statement is modified to obtain a real-time comment statement.
  • the device further includes: a statistics module 34 and an interaction module 35.
  • the statistics module 34 is configured to collect, according to the game data of the current game, a list of information under each action topic involved in the current game in real time;
  • the interaction module 35 is configured to receive a game question sent by the viewer during the current game, obtain an answer from the corresponding information list according to the action theme involved in the game question, and output an answer to the part of the game question in the interaction time of the current game. .
  • the device for performing live text broadcast on the game in this embodiment receives the game data in the current game in real time, and generates an emotion describing the game data in the current game according to the historical live speech with emotional sentiment describing the game data in the historical game.
  • the live broadcast statement tends to output the live broadcast statement to achieve live broadcast of the text of the current game. Since the device of the embodiment adopts a live speech with emotional tendency, the effect of the live broadcast of the current game has a swaying effect, and the quality of the live broadcast of the game is improved.
  • FIG. 5 is a schematic structural diagram of an apparatus for processing a live game statement according to an embodiment of the present disclosure. As shown in FIG. 5, the apparatus includes an acquisition module 51 and a marker scoring module 52.
  • the obtaining module 51 is configured to obtain a historical live speech statement with an emotional tendency used when the game data in the historical game is played live.
  • the tag scoring module 52 is configured to perform topic tagging and emotional scoring on the historical live speech according to the game action and the action result described by the historical live speech obtained by the obtaining module 51.
  • the obtaining module 51 is specifically configured to: obtain a live broadcast statement used when the text is live broadcasted by the artificial typing, as a historical live broadcast statement. or
  • the obtaining module 51 is specifically configured to: obtain a live broadcast sentence obtained by text conversion of audio data generated by performing live broadcast of a historical game by using an audio live broadcast method, as the historical live broadcast statement. or
  • the obtaining module 51 is specifically configured to: obtain a live broadcast sentence obtained by performing text conversion on the video data generated by the live broadcast of the historical game by using a live video method, as the historical live broadcast statement.
  • the marker scoring module 52 is specifically configured to:
  • the action descriptor, the action direction descriptor and the action degree descriptor are parsed from the historical live speech
  • the emotional score of the historical live speech is determined according to the action direction and the action degree described by the action direction descriptor.
  • the device for processing a live game statement in the embodiment is used for performing topic mark and emotional scoring on the historical live speech, and clearing the action theme and the emotional score of the live broadcast sentence, and texting the game for the device that performs live text on the game.
  • the live broadcast provides conditions so that the device that broadcasts the game live can generate a lively statement with an emotional tendency to broadcast the current game.
  • the disclosed system, apparatus, and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
  • the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
  • the software functional unit described above is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods described in various embodiments of the present application. Part of the steps.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Library & Information Science (AREA)
  • Machine Translation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请提供一种对比赛进行文字直播的方法及装置。对比赛进行文字直播方法包括：实时接收当前比赛中的比赛数据；根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句；输出实时直播语句。本申请可以抑扬顿挫的对比赛进行文字直播，提高了对比赛进行文字直播的质量。

Description

对比赛进行文字直播的方法及装置
本申请要求2015年08月18日递交的申请号为201510507932.X、发明名称为“对比赛进行文字直播的方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及互联网技术领域，尤其涉及一种对比赛进行文字直播的方法及装置。
背景技术
随着互联网技术的发展，出现了能够在用户终端(例如手机、平板电脑等)上对体育比赛进行文字直播的应用，例如虎扑看球APP。目前文字直播有两种方式，一种是基于人工打字的文字直播；一种是基于机器的文字直播。其中，相比于基于人工打字的文字直播方法，基于机器的文字直播方法能够节约人力资源，直播效率较高，因此被广泛采用。但是，基于机器的文字直播方法仅仅是直接对比赛进行事实陈述，例如，2015-05-2810:15:19，勇士-火箭，库里，三分投篮，进球，目前这种直播方法比较平淡枯燥，导致直播的比赛几乎没有用户观看，直播质量较差。
发明内容
本申请的多个方面提供一种对比赛进行文字直播的方法及装置，用以抑扬顿挫的对比赛进行文字直播，提高对比赛进行文字直播的质量。
本申请的一方面，提供一种对比赛进行文字直播的方法，包括：
实时接收当前比赛中的比赛数据；
根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述所述当前比赛中的比赛数据的具有情感倾向的实时直播语句；
输出所述实时直播语句。
本申请的另一方面，提供一种对比赛直播语句进行处理的方法，包括：
获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句；
根据所述历史直播语句描述的比赛动作和动作结果，对所述历史直播语句进行主题标记和情感打分。
本申请的又一方面，提供一种对比赛进行文字直播的装置，包括：
接收模块，用于实时接收当前比赛中的比赛数据；
生成模块，用于根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述所述当前比赛中的比赛数据的具有情感倾向的实时直播语句；
输出模块，用于输出所述实时直播语句。
本申请的又一方面，提供一种对比赛直播语句进行处理的装置，包括：
获取模块，用于获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句；
标记打分模块，用于根据所述历史直播语句描述的比赛动作和动作结果，对所述历史直播语句进行主题标记和情感打分。
在本申请中，实时接收当前比赛中的比赛数据，依据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句，输出该实时直播语句，实现对当前比赛的文字直播。由于本申请采用的是具有情感倾向的直播语句，所以对当前比赛的文字直播具有抑扬顿挫的效果，提高了对比赛进行文字直播的质量。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本申请一实施例提供的对比赛进行文字直播的方法的流程示意图；
图2为本申请一实施例提供的对比赛直播语句进行处理的方法的流程示意图；
图3为本申请一实施例提供的对比赛进行文字直播的装置的结构示意图；
图4为本申请另一实施例提供的对比赛进行文字直播的装置的结构示意图；
图5为本申请一实施例提供的对比赛直播语句进行处理的装置的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例 是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
在现有技术中，基于机器的文字直播方法仅仅是对比赛进行事实陈述，例如，2015-05-28 10:15:19，勇士-火箭，库里，三分投篮，进球，直播效果在情感上比较平淡枯燥，导致直播的比赛几乎没有用户观看，直播质量较差。针对该问题，本申请提供一种对比赛进行文字直播的方法，该方法属于基于机器的文字直播方法，具体原理是：基于描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句，实现对当前比赛的文字直播。由于本申请在对当前比赛进行文字直播时采用的是具有情感倾向的直播语句，所以对当前比赛的文字直播具有抑扬顿挫的效果，提高了对比赛进行文字直播的质量。
图1为本申请一实施例提供的对比赛进行文字直播的方法的流程示意图。如图1所示，该方法包括：
101、实时接收当前比赛中的比赛数据。
102、根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句。
103、输出该实时直播语句。
在本实施例中，实时获取当前比赛中的比赛数据，用以对当前比赛进行实时直播。本实施例中当前比赛的比赛数据可以实时从比赛方接入，可由官方提供。在本实施例中，无论是当前比赛中的比赛数据还是历史比赛中的比赛数据，所述比赛数据主要包括：比赛时间、比赛方、比赛人物、比赛动作、动作结果等信息。例如，在勇士对火箭的比赛中的部分比赛数据如下：
2015-05-28 10:15:13，勇士-火箭，库里，传球，汤普森接球；
2015-05-28 10:15:16，勇士-火箭，汤普森，传球，库里接球；
2015-05-28 10:15:19，勇士-火箭，库里，三分投篮，进球。
由上述可见，上述比赛数据仅仅是简单的叙述比赛事实，没有情感波动，如果采用这种比赛数据进行文字直播，直播质量较差，将无法吸引观众观看。
于是，本实施例根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成具有情感倾向的实时直播语句，采用具有情感倾向的实时直播语句对当前比赛进行文字直播，使得直播效果具有情感上的抑扬顿挫，吸引观众观看，提高文字直播的质量， 提升观众的观看体验等。
上述历史直播语句可以是采用人工的文字直播方法对历史比赛进行文字直播时使用的直播语句，或者可以是由采用音频直播方法对历史比赛进行直播产生的音频数据进行文本转换后得到的直播语句，或者可以是由采用视频直播方法对历史比赛进行直播产生的视频数据进行文本转换后得到的直播语句。其中，采用人工的文字直播方法对比赛进行文字直播时使用的直播语句或者由音频、视频直播产生的数据进行文本转换得到的直播语句一般具有情感倾向，直播质量较好，如下所示：
2015-05-28 10:15:13，勇士-火箭，“库里将球传给汤普森”
2015-05-28 10:15:16，勇士-火箭，“汤普森运了下球后，回传给库里”
2015-05-28 10:15:15，勇士-火箭，“库里接球不看筐三分球出手”
2015-05-28 10:15:17，勇士-火箭，“有木有！！！”
2015-05-28 10:15:19，勇士-火箭，“OMG！！！空心！！！”
2015-05-28 10:15:17，勇士-火箭，“库里再次向我们展示了什么是MVP！！！”
值得说明的是，人工的文字直播方法需要耗费较多人力资源，实现成本较高。采用本实施例提供的方法，可以实现具有情感倾向的文字直播，且可以解决人力资源耗费等问题。另外，在本实施例提供的方法被实施之后，当采用本实施例对当前比赛进行文字直播时，还可以获取之前采用本实施例方法进行文字直播时使用的具有情感倾向的直播语句作为历史直播语句。
在一可选实施方式中，为了便于根据具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句，本实施例还提供一种对比赛直播语句进行处理方法，用以预先对历史直播语句进行主题标记和情感打分。如图2所示，该方法包括：
201、获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句。
例如，获取采用人工打字的文字直播方法对历史比赛进行文字直播时使用的直播语句，作为这里的历史直播语句。或者
获取由采用音频直播方法对历史比赛进行直播产生的音频数据进行文本转换后得到的直播语句，作为这里的历史直播语句；或者
获取由采用视频直播方法对所述历史比赛进行直播产生的视频数据进行文本转换后得到的直播语句，作为这里的历史直播语句。
202、根据历史直播语句描述的比赛动作和动作结果，对历史直播语句进行主题标记和情感打分。
在本实施例中，具体根据当前比赛涉及的比赛动作设定若干动作主题，以篮球比赛为例，一般包括“运球”、“传球”、“上篮”、“扣篮”、“篮板”、“盖帽”等动作主题。其中，每个动作主题需要利用一些关键词去描述，进而根据这些关键词，可以确定每个动作主题下对应的直播语句。
对历史直播语句进行情感打分的目的是为了区分不同历史直播语句在情感上的差异，情感得分不同的历史直播语句表达的情感不同。其中，历史直播语句的情感得分可以由历史直播语句描述的动作方向和程度来确定，所述动作方向实际上是指动作结果，例如投篮投中与否，或者运球流畅与否等。
基于上述，在对历史直播语句进行主题标记和情感打分时，具体可以：从历史直播语句中解析出动作描述词、动作方向描述词和动作程度描述词；根据动作描述词所描述的比赛动作，标记历史直播语句的动作主题；根据动作方向描述词所描述的动作方向和动作程度描述词所描述的动作程度，确定所述历史直播语句的情感得分。
例如，假设历史直播语句为：库里运球很流畅，则从该历史直播语句中解析出动作描述词、动作方向描述词和动作程度描述词分别为：运球，流畅，很。其中，根据动作描述词可知该历史直播语句涉及的比赛动作为运球，标记该历史直播语句的动作主题为：运球；进一步，根据动作方向描述词所描述的动作方向以及动作程度描述词所描述的动作程度，对历史直播语句进行情感打分。其中，可以预先设定根据动作方向描述词所描述的动作方向以及动作程度描述词所描述的动作程度，对历史直播语句进行情感打分的策略或规则，基于所设定的策略或规则，对历史直播语句进行情感打分。
以对历史直播语句进行-5到+5的情感打分为例，假设给积极的动作方向打正分，根据动作程度打+3到+5分，例如运球流畅打正分，根据流畅程度可以打+3到+5分，这里给“很流畅”打+4分；假设给消极的动作方向打负分，根据动作程度打-3到-5分，例如给运球不流畅打负分，根据不流畅程度可以打-3到-5分，若是“很不流畅”则可以打-4分；等等。一种对历史直播语句进行主题标记和情感打分后的结果如下所示：
2015-05-28 10:15:13，勇士-火箭，“库里将球传给汤普森”，传球，3分
2015-05-28 10:15:16，勇士-火箭，“汤普森运了下球后，回传给库里”，传球，3分
2015-05-28 10:15:19，勇士-火箭，“库里接球不看筐三分球出手”，投篮，4分
2015-05-28 10:15:21，勇士-火箭，“有木有！！！”，投篮，4分
2015-05-28 10:15:23，勇士-火箭，“OMG！！！空心！！！”，进球，5分
2015-05-28 10:15:25，勇士-火箭，“库里再次向我们展示了什么是MVP！！！”，进球类评论，5分
在历史直播语句后面的“传球”、“投篮”等词表示动作主题，“3分”、“4分”等表示历史直播语句的情感得分。这里需要注意的是“进球类评论”属于评论语句，在后续会有详细描述。
值得说明的是，为了便于使用，可以将主题标记和情感打分后的历史直播语句形成文字直播列表并输出，以供后续对比赛进行文字直播使用。
在对历史直播语句进行主题标记和情感打分的基础上，根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句的实施方式包括：
确定当前比赛中的比赛数据描述的比赛动作和动作结果；
根据当前比赛中的比赛数据描述的比赛动作和动作结果以及历史直播语句的动作主题和情感得分，从历史直播语句中获得情感与当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句；
对候选历史直播语句进行修正，以获得实时直播语句。
其中，可以设定策略或规则。按照设定的策略或规则，根据当前比赛中的比赛数据描述的比赛动作和动作结果以及历史直播语句的动作主题和情感得分，从历史直播语句中获得情感与当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句。
例如，在一种规则中，假设传球类动作适用于传球主题中情感得分1-3的直播语句；进球类动作适用于投篮主题中情感得分3～5的直播语句；若不进球，则适用于投篮主题中情感得分-3～-5的直播语句，等等。则按照该规则，就可以根据当前比赛中的比赛数据描述的比赛动作和动作结果以及历史直播语句的动作主题和情感得分，从历史直播语句中获得情感与当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句。
在获得候选历史直播语句之后，候选历史直播语句可能不能直接用作实时直播语句，例如候选历史直播语句涉及的比赛人物可能与当前比赛的比赛数据中的比赛人物不同，所以可以对候选历史直播语句进行修正，将候选历史直播语句中比赛人物替换为当前比赛中的比赛人物。又或者，可以对候选历史直播语句进行表述上的调整，例如减少一个感叹号，删除几个字词等，或适当增加几个字词等，以获得更加适合当前比赛的实时直 播语句。
在一可选实施方式中，可能会获取到多个情感与当前比赛的比赛数据相符的候选历史直播语句，则可以从多个候选历史直播语句中，选择情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最小的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句，作为目标历史直播语句，对该目标历史直播语句进行修正，以获得实时直播语句。
例如，对于像进球类动作等涉及正向情感的候选历史直播语句，可以选择情感得分的绝对值最高的候选历史直播语句，例如从得分为3～5的候选历史直播语句中选择得分为5的候选历史直播语句，作为目标历史直播语句。
对于像不进球类动作等涉及负向情感的候选历史直播语句，可以选择情感得分的绝对值最高的候选历史直播语句，例如从得分为-3～-5的候选历史直播语句中选择得分为-5的候选历史直播语句，作为目标历史直播语句。
对于像传球类动作等涉及中性情感的候选历史直播语句，可以选择情感得分的绝对值最低的候选历史直播语句，例如从得分为-2～2的候选历史直播语句中选择得分为0的候选历史直播语句，作为目标历史直播语句。
若情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最小的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句为多个，则可以从中随机选择一个作为目标历史直播语句。
若只有一个候选历史直播语句，则可以直接对该候选历史直播语句进行修正，以获得实时直播语句。
由上述可见，通过采用具有情感倾向的实时直播语句对当前比赛进行文字直播，可以实现抑扬顿挫的直播效果，提高了文字直播效率，有利于吸引观众观看直播比赛。
进一步，本申请的提供的方法除了用上述具有情感倾向的直播语句对当前比赛进行直播外，还可以实时对比赛进行评论。
具体的，还可以根据对历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对当前比赛进行实时评论的具有情感倾向的实时评论语句；输出该实时评论语句，以对比赛进行实时评论，进一步提高直播质量，提高用户对直播比赛的兴趣，吸引观众观看直播比赛。
一种生成实时评论语句的方式为：
确定当前比赛中待评论的比赛内容；
根据该比赛内容涉及的比赛动作和动作结果以及历史评论语句的动作主题和情感得分，从历史评论语句中获得情感与该比赛内容相符的历史评论语句作为候选历史评论语句；
对候选历史评论语句进行修正，以获得实时评论语句。
其中，在确定当前比赛中待评论的比赛内容过程中，可以预先设定进行评论的时间点，在该时间点之前播出的比赛数据构成待评论的比赛内容；或者可以预先设定需要进行评论的比赛情景，例如每次进球时，或者每次失球时，或者某一队连续进球时，或者某一队连续失球时等，则可以将与进球或失球有关的比赛数据作为待评论内容等；又或者，可以设定指定的比赛人物，例如汤普森，当出现与该指定比赛人物有关的比赛数据时，将与该比赛人物有关的信息作为待评论的比赛内容。
同理，在选择候选历史评论语句过程中，也可以设定策略或规则。则按照预设的策略或规则，根据该比赛内容涉及的比赛动作和动作结果以及历史评论语句的动作主题和情感得分，从历史评论语句中获得情感与该比赛内容相符的历史评论语句作为候选历史评论语句。
例如，假设将某一球队连续“进球”或者连续“不进球”时进行评论，待评论的比赛内容是与该连续“进球”或者连续“不进球”有关的内容。对于一支球队若连续进球，其情感分就会很高，则一种策略或规则可以是根据当前比赛情况选择对应正面情感得分3～5的历史评论语句；若一一支球队进球不连续，其情感比较一般，属于中性情感，则获取情感得分-2～2的；历史评论语句；若一支球队连续不进球，其负面情感会比较高，于是可以选择情感得分-3～-5的历史评论语句。由此可见，具体选择何种历史评论语句实际上是由待评论的比赛内容以及该比赛内容涉及的动作主题以及全局情感共同决定的。
例如，在上述举例中，“2015-05-28 10:15:25，勇士-火箭，“库里再次向我们展示了什么是MVP！！！”，进球类评论，5分”属于文字直播中的评论语句，该评论语句具体涉及进球类评论。对于历史评论语句的动作主题和情感得分，也可以通过主题标记和情感打分来获得。
由此可见，本实施例通过具有情感倾向的直播语句对当前比赛进行直播，而且通过具有情感的评论语句对当前比赛进行实时评论，有利于进一步提高文字直播的质量，吸引更多观众观看直播比赛。
更进一步，本实施例提供的方法还包括：根据当前比赛的比赛数据，实时统计当前 比赛涉及的每个动作主题下的信息列表；接收观众在当前比赛过程中发送的比赛问题，根据比赛为涉及的动作主题从相应信息列表中获取答案，并在当前比赛的互动时间，输出比赛结果中部分比赛结果的答案，实现与观众的互动，并实时解答观众在比赛中提出的问题。
实时接收到的当前比赛的比赛数据可以存储在直播数据库中。在直播数据库中存储有与当前比赛有关的全量比赛数据，不同比赛有不同类型的数据。具体实现方法如下：根据当前比赛涉及的动作主题对直播数据库中的比赛数据分类建表，例如对于投篮这个主题，可以用一张信息列表来存储各个球员在比赛中的命中率、失误率等，对于传球这个主题，可以用一张信息列表来存储各球员在比赛中的传球次数等信息，以便于解答观众提出的问题时查找答案。
例如，假设观众在当前比赛过程中提出的比赛问题是：“库里到现在为止命中率多少？”。根据该比赛问题，可以解析观众提出的比赛问题，确定该问题涉及投篮主题，于是去后台信息列表中搜索相应结果。比如，对于比赛问题“库里到现在为止命中率多少？”，找到的答案是“50％”。其中，该搜索答案的过程在技术实现上可以利用Lucene、Solr等成熟的搜索引擎架构来实现，实时计算可以应用storm流计算框架，但不限于此。
其中，与观众的互动通常发生在比赛暂停或者中场休息时，意味着当前比赛的互动时间是指比赛暂停或中场休息时间。由于观众互动时间往往有限，而观众提出的比赛问题数量会很多，一般无法答复所有比赛问题，因此可以适应性的选择部分比赛问题进行答复。一种最基本的选择方式是：选择最热门的比赛问题给予展示。例如，可以此阿勇Word2Vec模型，将对比赛问题以及答案进行聚类，选择互动最多的那个类别，从中抽取一个或几个比赛问题进行解答。根据互动时间长短，可以按热度对不同数量的比赛问题进行解答。
由此可见，本申请提供的对比赛进行文字直播的方法，利用具有情感倾向的直播语句对当前比赛进行直播，并利用具有情感倾向的评论语句对当前比赛进行实时评论，同时还支持与观众互动，解答观众提出的比赛问题，提高文字直播的质量，提高了观众收看直播的体验，能够基本摆脱人工直播，降低了成本，提高了效率。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定 是本申请所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
图3为本申请一实施例提供的对比赛进行文字直播的装置的结构示意图。如图3所示，该装置包括：接收模块31、生成模块32和输出模块33。
接收模块31，用于实时接收当前比赛中的比赛数据。
生成模块32，用于根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述接收模块31接收的当前比赛中的比赛数据的具有情感倾向的实时直播语句。
输出模块33，用于输出生成模块32生成的实时直播语句。
在一可选实施方式中，生成模块32具体可用于：
确定当前比赛中的比赛数据描述的比赛动作和动作结果；
根据当前比赛中的比赛数据描述的比赛动作和动作结果以及历史直播语句的动作主题和情感得分，从历史直播语句中获得情感与当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句；
对候选历史直播语句进行修正，以获得实时直播语句。
进一步，生成模块32在对候选历史直播语句进行修正，以获得实时直播语句时，具体可用于：
若获得多个候选历史直播语句，从多个候选历史直播语句中，选择情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最低的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句作为目标历史直播语句；
对目标历史直播语句进行修正，以获得所述实时直播语句。
例如，对于像进球类动作等涉及正向情感的候选历史直播语句，可以选择情感得分的绝对值最高的候选历史直播语句，例如从得分为3～5的候选历史直播语句中选择得分为5的候选历史直播语句，作为目标历史直播语句。
对于像不进球类动作等涉及负向情感的候选历史直播语句，可以选择情感得分的绝对值最高的候选历史直播语句，例如从得分为-3～-5的候选历史直播语句中选择得分为-5的候选历史直播语句，作为目标历史直播语句。
对于像传球类动作等涉及中性情感的候选历史直播语句，可以选择情感得分的绝对值最低的候选历史直播语句，例如从得分为-2～2的候选历史直播语句中选择得分为0的候选历史直播语句，作为目标历史直播语句。
若情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最小的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句为多个，则可以从中随机选择一个作为目标历史直播语句。
在一可选实施方式中，生成模块32还用于：根据对历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对当前比赛进行实时评论的具有情感倾向的实时评论语句。相应的，输出模块33还用于：输出生成模块32生成的实时评论语句。
可选的，生成模块32在根据对历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对当前比赛进行实时评论的具有情感倾向的实时评论语句时，具体可用于：
确定当前比赛中待评论的比赛内容；
根据比赛内容涉及的比赛动作和动作结果以及历史评论语句的动作主题和情感得分，从历史评论语句中获得情感与该比赛内容相符的历史评论语句作为候选历史评论语句；
对候选历史评论语句进行修正，以获得实时评论语句。
在一可选实施方式中，如图4所示，该装置还包括：统计模块34和互动模块35。
统计模块34，用于根据当前比赛的比赛数据，实时统计当前比赛涉及的每个动作主题下的信息列表；
互动模块35，用于接收观众在当前比赛过程中发送的比赛问题，根据比赛问题涉及的动作主题从相应信息列表中获取答案，并在当前比赛的互动时间，输出比赛问题中部分比赛问题的答案。
本实施例提供的对比赛进行文字直播的装置，实时接收当前比赛中的比赛数据，依据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述当前比赛中的比赛数据的具有情感倾向的实时直播语句，输出该实时直播语句，实现对当前比赛的文字直播。由于本实施例装置采用的是具有情感倾向的直播语句，所以对当前比赛的文字直播具有抑扬顿挫的效果，提高了对比赛进行文字直播的质量。
图5为本申请一实施例提供的对比赛直播语句进行处理的装置的结构示意图。如图5所示，该装置包括：获取模块51和标记打分模52。
获取模块51，用于获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句。
标记打分模块52，用于根据获取模块51获取的历史直播语句描述的比赛动作和动作结果，对历史直播语句进行主题标记和情感打分。
在一可选实施方式中，获取模块51具体用于：获取采用人工打字的文字直播方法对历史比赛进行文字直播时使用的直播语句，作为历史直播语句。或者
获取模块51具体用于：获取由采用音频直播方法对历史比赛进行直播产生的音频数据进行文本转换后得到的直播语句，作为所述历史直播语句。或者
获取模块51具体用于：获取由采用视频直播方法对历史比赛进行直播产生的视频数据进行文本转换后得到的直播语句，作为所述历史直播语句。
在一可选实施方式中，标记打分模块52具体用于：
从历史直播语句中解析出动作描述词、动作方向描述词和动作程度描述词；
根据动作描述词所描述的比赛动作，标记历史直播语句的动作主题；
根据动作方向描述词所描述的动作方向和动作程度描述词的动作程度，确定历史直播语句的情感得分。
本实施例提供的对比赛直播语句进行处理的装置，用于对历史直播语句进行主题标记和情感打分，明确历史直播语句的动作主题和情感得分，为对比赛进行文字直播的装置对比赛进行文字直播提供条件，以便于对比赛进行文字直播的装置能够生成具有情感倾向的直播语句对当前比赛进行直播。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (18)

  1. 一种对比赛进行文字直播的方法，其特征在于，包括：
    实时接收当前比赛中的比赛数据；
    根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述所述当前比赛中的比赛数据的具有情感倾向的实时直播语句；
    输出所述实时直播语句。
  2. 根据权利要求1所述的方法，其特征在于，所述根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句，生成描述所述当前比赛中的比赛数据的具有情感倾向的实时直播语句，包括：
    确定所述当前比赛中的比赛数据描述的比赛动作和动作结果；
    根据所述当前比赛中的比赛数据描述的比赛动作和动作结果以及所述历史直播语句的动作主题和情感得分，从所述历史直播语句中获得情感与所述当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句；
    对所述候选历史直播语句进行修正，以获得所述实时直播语句。
  3. 根据权利要求2所述的方法，其特征在于，所述对所述候选历史直播语句进行修正，以获得所述实时直播语句，包括：
    若获得多个候选历史直播语句，从所述多个候选历史直播语句中，选择情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最低的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句作为目标历史直播语句；
    对所述目标历史直播语句进行修正，以获得所述实时直播语句。
  4. 根据权利要求1所述的方法，其特征在于，还包括：
    根据对所述历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对所述当前比赛进行实时评论的具有情感倾向的实时评论语句；
    输出所述实时评论语句。
  5. 根据权利要求4所述的方法，其特征在于，所述根据对所述历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对所述当前比赛进行实时评论的具有情感倾向的实时评论语句，包括：
    确定所述当前比赛中待评论的比赛内容；
    根据所述比赛内容涉及的比赛动作和动作结果以及所述历史评论语句的动作主题和情感得分，从所述历史评论语句中获得情感与所述比赛内容相符的历史评论语句作为 候选历史评论语句；
    对所述候选历史评论语句进行修正，以获得所述实时评论语句。
  6. 根据权利要求1-5任一项所述的方法，其特征在于，还包括：
    根据所述当前比赛的比赛数据，实时统计所述当前比赛涉及的每个动作主题下的信息列表；
    接收观众在所述当前比赛过程中发送的比赛问题，根据所述比赛问题涉及的动作主题从相应信息列表中获取答案；
    在所述当前比赛的互动时间，输出所述比赛问题中部分比赛问题的答案。
  7. 一种对比赛直播语句进行处理的方法，其特征在于，包括：
    获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句；
    根据所述历史直播语句描述的比赛动作和动作结果，对所述历史直播语句进行主题标记和情感打分。
  8. 根据权利要求7所述的方法，其特征在于，所述获取对历史比赛中的比赛数据进行直播时使用的具有情感倾向的历史直播语句，包括：
    获取采用人工打字的文字直播方法对所述历史比赛进行文字直播时使用的直播语句，作为所述历史直播语句；或者
    获取由采用音频直播方法对所述历史比赛进行直播产生的音频数据进行文本转换后得到的直播语句，作为所述历史直播语句；或者
    获取由采用视频直播方法对所述历史比赛进行直播产生的视频数据进行文本转换后得到的直播语句，作为所述历史直播语句。
  9. 根据权利要求7或8所述的方法，其特征在于，所述根据所述历史直播语句描述的比赛动作和动作结果，对所述历史直播语句进行主题标记和情感打分，包括：
    从所述历史直播语句中解析出动作描述词、动作方向描述词和动作程度描述词；
    根据所述动作描述词所描述的比赛动作，标记所述历史直播语句的动作主题；
    根据所述动作方向描述词所描述的动作方向和所述动作程度描述词的动作程度，确定所述历史直播语句的情感得分。
  10. 一种对比赛进行文字直播的装置，其特征在于，包括：
    接收模块，用于实时接收当前比赛中的比赛数据；
    生成模块，用于根据描述历史比赛中的比赛数据的具有情感倾向的历史直播语句， 生成描述所述当前比赛中的比赛数据的具有情感倾向的实时直播语句；
    输出模块，用于输出所述实时直播语句。
  11. 根据权利要求10所述的装置，其特征在于，所述生成模块具体用于：
    确定所述当前比赛中的比赛数据描述的比赛动作和动作结果；
    根据所述当前比赛中的比赛数据描述的比赛动作和动作结果以及所述历史直播语句的动作主题和情感得分，从所述历史直播语句中获得情感与所述当前比赛中的比赛数据相符的历史直播语句作为候选历史直播语句；
    对所述候选历史直播语句进行修正，以获得所述实时直播语句。
  12. 根据权利要求11所述的装置，其特征在于，所述生成模块具体用于：
    若获得多个候选历史直播语句，从所述多个候选历史直播语句中，选择情感得分的绝对值最高的候选历史直播语句或者情感得分的绝对值最低的候选历史直播语句或者文本表达符合预设要求的候选历史直播语句作为目标历史直播语句；
    对所述目标历史直播语句进行修正，以获得所述实时直播语句。
  13. 根据权利要求10所述的装置，其特征在于，
    所述生成模块还用于：
    根据对所述历史比赛进行实时评论的具有情感倾向的历史评论语句，生成对所述当前比赛进行实时评论的具有情感倾向的实时评论语句；
    所述输出模块还用于：
    输出所述实时评论语句。
  14. 根据权利要求13所述的装置，其特征在于，所述生成模块具体用于：
    确定所述当前比赛中待评论的比赛内容；
    根据所述比赛内容涉及的比赛动作和动作结果以及所述历史评论语句的动作主题和情感得分，从所述历史评论语句中获得情感与所述比赛内容相符的历史评论语句作为候选历史评论语句；
    对所述候选历史评论语句进行修正，以获得所述实时评论语句。
  15. 根据权利要求10-14任一项所述的装置，其特征在于，还包括：
    统计模块，用于根据所述当前比赛的比赛数据，实时统计所述当前比赛涉及的每个动作主题下的信息列表；
    互动模块，用于接收观众在所述当前比赛过程中发送的比赛问题，根据所述比赛问题涉及的动作主题从相应信息列表中获取答案，并在所述当前比赛的互动时间，输出所 述比赛问题中部分比赛问题的答案。
  16. 一种对比赛直播语句进行处理的装置，其特征在于，包括：
    获取模块，用于获取对历史比赛中的比赛数据进行文字直播时使用的具有情感倾向的历史直播语句；
    标记打分模块，用于根据所述历史直播语句描述的比赛动作和动作结果，对所述历史直播语句进行主题标记和情感打分。
  17. 根据权利要求16所述的装置，其特征在于，所述获取模块具体用于：
    获取采用人工打字的文字直播方法对所述历史比赛进行文字直播时使用的直播语句，作为所述历史直播语句；或者
    获取由采用音频直播方法对所述历史比赛进行直播产生的音频数据进行文本转换后得到的直播语句，作为所述历史直播语句；或者
    获取由采用视频直播方法对所述历史比赛进行直播产生的视频数据进行文本转换后得到的直播语句，作为所述历史直播语句。
  18. 根据权利要求16或17所述的装置，其特征在于，所述标记打分模块具体用于：
    从所述历史直播语句中解析出动作描述词、动作方向描述词和动作程度描述词；
    根据所述动作描述词所描述的比赛动作，标记所述历史直播语句的动作主题；
    根据所述动作方向描述词所描述的动作方向和所述动作程度描述词的动作程度，确定所述历史直播语句的情感得分。
PCT/CN2016/094115 2015-08-18 2016-08-09 对比赛进行文字直播的方法及装置 WO2017028706A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US15/752,956 US10474749B2 (en) 2015-08-18 2016-08-09 Method and device for generating live text broadcasting content using historical broadcasting text
JP2018507650A JP2018525675A (ja) 2015-08-18 2016-08-09 過去のブロードキャストテキストを用いてライブテキストブロードキャストコンテンツを生成する方法およびデバイス
SG11201801334TA SG11201801334TA (en) 2015-08-18 2016-08-09 Method and device for live text broadcasting of match
KR1020187007489A KR20180041189A (ko) 2015-08-18 2016-08-09 경기의 라이브 문자 방송을 위한 방법 및 디바이스

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510507932.XA CN106470363B (zh) 2015-08-18 2015-08-18 对比赛进行文字直播的方法及装置
CN201510507932.X 2015-08-18

Publications (1)

Publication Number Publication Date
WO2017028706A1 true WO2017028706A1 (zh) 2017-02-23

Family

ID=58050407

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/094115 WO2017028706A1 (zh) 2015-08-18 2016-08-09 对比赛进行文字直播的方法及装置

Country Status (6)

Country Link
US (1) US10474749B2 (zh)
JP (1) JP2018525675A (zh)
KR (1) KR20180041189A (zh)
CN (1) CN106470363B (zh)
SG (1) SG11201801334TA (zh)
WO (1) WO2017028706A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11159856B2 (en) * 2017-07-10 2021-10-26 Sony Interactive Entertainment LLC Non-linear content presentation and experience
CN107485843B (zh) * 2017-09-26 2018-07-20 北京球友圈网络科技有限责任公司 统计球类体育比赛信息的方法及装置
CN109145733A (zh) * 2018-07-17 2019-01-04 焦点科技股份有限公司 一种篮球比赛的人工智能解说方法及系统
US10805029B2 (en) * 2018-09-11 2020-10-13 Nbcuniversal Media, Llc Real-time automated classification system
JP7142315B2 (ja) * 2018-09-27 2022-09-27 パナソニックＩｐマネジメント株式会社 説明支援装置および説明支援方法
JP7254359B2 (ja) * 2020-06-08 2023-04-10 株式会社Ales プログラム、装置、および、脚本の製造方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10322233A (ja) * 1997-05-20 1998-12-04 Sony Corp 文字多重放送の受信機
KR20090075426A (ko) * 2008-01-04 2009-07-08 주식회사 대우일렉트로닉스 디지털 방송 수신기에서 문자방송 수신 방법
CN101730008A (zh) * 2008-10-20 2010-06-09 诺基亚公司 用于消息生成的方法、系统和装置
CN101795287A (zh) * 2009-12-29 2010-08-04 上海酷吧信息技术有限公司 移动终端实现比赛文字直播的方法
CN102929861A (zh) * 2012-10-22 2013-02-13 杭州东信北邮信息技术有限公司 一种文本情感指数计算方法和系统

Family Cites Families (36)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005099423A2 (en) 2004-04-16 2005-10-27 Aman James A Automatic event videoing, tracking and content generation system
US20150297949A1 (en) 2007-06-12 2015-10-22 Intheplay, Inc. Automatic sports broadcasting system
US6976031B1 (en) 1999-12-06 2005-12-13 Sportspilot, Inc. System and method for automatically generating a narrative report of an event, such as a sporting event
US7333967B1 (en) 1999-12-23 2008-02-19 International Business Machines Corporation Method and system for automatic computation creativity and specifically for story generation
US6810397B1 (en) * 2000-06-29 2004-10-26 Intel Corporation Collecting event data and describing events
US20030056116A1 (en) 2001-05-18 2003-03-20 Bunker Nelson Waldo Reporter
JP2002239225A (ja) * 2001-11-12 2002-08-27 Namco Ltd ゲーム実況中継装置
JP3789854B2 (ja) * 2002-06-11 2006-06-28 エヌ・ティ・ティ・コムウェア株式会社 ライブ配信サーバ、及びライブ配信方法
US7593687B2 (en) 2003-10-07 2009-09-22 Immersion Entertainment, Llc System and method for providing event spectators with audio/video signals pertaining to remote events
JP2005191892A (ja) * 2003-12-25 2005-07-14 Sharp Corp 情報取得装置及びこれを用いたマルチメディア情報作成システム
US7933958B2 (en) 2004-06-08 2011-04-26 International Business Machines Corporation Real-time blog interaction
JP4981026B2 (ja) 2005-03-31 2012-07-18 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 複合ニュース・ストーリーの合成
US20080043089A1 (en) 2006-06-16 2008-02-21 Auerbach Philip Real time interactive entertainment
US7856390B2 (en) 2007-06-06 2010-12-21 Vhs, Llc System, report, and method for generating natural language news-based stories
US8494944B2 (en) 2007-06-06 2013-07-23 O2 Media, LLC System, report, and method for generating natural language news-based stories
US8136133B2 (en) 2007-11-13 2012-03-13 Walker Digital, Llc Methods and systems for broadcasting modified live media
US20090326948A1 (en) 2008-06-26 2009-12-31 Piyush Agarwal Automated Generation of Audiobook with Multiple Voices and Sounds from Text
US8307086B2 (en) 2008-08-19 2012-11-06 Facebook, Inc. Resource management of social network applications
US20100132049A1 (en) 2008-11-26 2010-05-27 Facebook, Inc. Leveraging a social graph from a social network for social context in other systems
US8468153B2 (en) 2009-01-21 2013-06-18 Recorded Future, Inc. Information service for facts extracted from differing sources on a wide area network
CN102118580B (zh) * 2009-12-30 2015-02-18 新奥特（北京）视频技术有限公司 一种数据分组动态替换的字幕播出方法和装置
CN101730009A (zh) 2009-12-31 2010-06-09 宇龙计算机通信科技(深圳)有限公司 一种短信与邮件相互转发的方法、装置及移动终端
CN102196193B (zh) * 2010-03-15 2015-04-29 新奥特（北京）视频技术有限公司 一种实现多种球类体育赛事的字幕播出方法
WO2011127140A1 (en) 2010-04-06 2011-10-13 Statsheet, Inc. Systems for dynamically generating and presenting narrative content
US8374848B1 (en) 2010-05-13 2013-02-12 Northwestern University System and method for using data and derived features to automatically generate a narrative story
US8355903B1 (en) 2010-05-13 2013-01-15 Northwestern University System and method for using data and angles to automatically generate a narrative story
US8661076B2 (en) 2010-09-23 2014-02-25 Salesforce.Com, Inc. Business networking information feed alerts
WO2012135362A2 (en) * 2011-03-28 2012-10-04 Waze, Inc. Device, system and method of television broadcasting of live feed from mobile devices
JP2013229873A (ja) * 2012-04-24 2013-11-07 Dowango:Kk 候補選択装置、候補選択方法及びコンピュータプログラム
US20130346867A1 (en) 2012-06-25 2013-12-26 United Video Properties, Inc. Systems and methods for automatically generating a media asset segment based on verbal input
US20140082645A1 (en) 2012-09-14 2014-03-20 Peter Stern Apparatus and methods for providing enhanced or interactive features
US9509758B2 (en) 2013-05-17 2016-11-29 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Relevant commentary for media content
JP2015115892A (ja) * 2013-12-13 2015-06-22 シャープ株式会社 コメント作成装置およびその制御方法
CN104618806A (zh) * 2014-03-17 2015-05-13 腾讯科技（北京）有限公司 获取视频的评论信息的方法、装置及系统
CN104199933B (zh) * 2014-09-04 2017-07-07 华中科技大学 一种多模态信息融合的足球视频事件检测与语义标注方法
US10425700B2 (en) * 2016-12-31 2019-09-24 Turner Broadcasting System, Inc. Dynamic scheduling and channel creation based on real-time or near-real-time content context analysis

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10322233A (ja) * 1997-05-20 1998-12-04 Sony Corp 文字多重放送の受信機
KR20090075426A (ko) * 2008-01-04 2009-07-08 주식회사 대우일렉트로닉스 디지털 방송 수신기에서 문자방송 수신 방법
CN101730008A (zh) * 2008-10-20 2010-06-09 诺基亚公司 用于消息生成的方法、系统和装置
CN101795287A (zh) * 2009-12-29 2010-08-04 上海酷吧信息技术有限公司 移动终端实现比赛文字直播的方法
CN102929861A (zh) * 2012-10-22 2013-02-13 杭州东信北邮信息技术有限公司 一种文本情感指数计算方法和系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
QIAO, WEI;: "Research on the Language and Expression of Text Live Broadcast on Football on Domestic Portal Websites", GRADUATION THESIS FOR MASTER'S DEGREE, 20 May 2014 (2014-05-20) *

Also Published As

Publication number Publication date
CN106470363B (zh) 2019-09-13
SG11201801334TA (en) 2018-03-28
US10474749B2 (en) 2019-11-12
US20180239750A1 (en) 2018-08-23
JP2018525675A (ja) 2018-09-06
KR20180041189A (ko) 2018-04-23
CN106470363A (zh) 2017-03-01

Similar Documents

Publication Publication Date Title
WO2017028706A1 (zh) 对比赛进行文字直播的方法及装置
US11550998B2 (en) Method and apparatus for generating a competition commentary based on artificial intelligence, and storage medium
Maloney et al. ‘Mmm… I love it, bro!’: Performances of masculinity in YouTube gaming
WO2019228268A1 (zh) 直播间显示方法、装置、设备及存储介质
CN106933808A (zh) 基于人工智能的文章标题生成方法、装置、设备及介质
CN112015852A (zh) 在关于事件的会话中提供响应
CN108460122B (zh) 基于深度学习的视频搜索方法、存储介质、设备及系统
US11833433B2 (en) Online virtual narration method and device, and medium
Ehret et al. How speculative designs produce new potentials for education research in digital culture
Tanaka et al. Lol-v2t: Large-scale esports video description dataset
CN113497946A (zh) 视频处理方法、装置、电子设备和存储介质
Domeneghetti Insights on reporting sports in the digital age: ethical and practical considerations in a changing media landscape
CN112423093B (zh) 游戏视频生成方法、装置、服务器和存储介质
US20220062772A1 (en) Determining game quality based on gameplay duration
TWI735510B (zh) 對比賽進行文字直播的方法及裝置
CN109388806B (zh) 一种基于深度学习及遗忘算法的中文分词方法
WO2018072577A1 (zh) 一种文本生成方法及服务器
Liu et al. Video-csr: Complex video digest creation for visual-language models
CN105611419A (zh) 显示观众猜测比分的方法和装置
Xue The Transmission Effect of Costume Dramas on International Social Media
US20200175058A1 (en) Methods and systems generating curated playlists
CN113992962B (zh) 弹幕发布方法、装置、电子设备及存储介质
Abideen et al. Ball-by-Ball Cricket Commentary Generation using Stateful Sequence-to-Sequence Model
Kalima The effect of topic on word formation and the frequency of neologism use on Internet forums
CN105608623A (zh) 显示猜测比分的方法和装置

Legal Events

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

Ref document number: 16836570

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018507650

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 15752956

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 11201801334T

Country of ref document: SG

ENP Entry into the national phase

Ref document number: 20187007489

Country of ref document: KR

Kind code of ref document: A

122 Ep: pct application non-entry in european phase

Ref document number: 16836570

Country of ref document: EP

Kind code of ref document: A1