CN108536802A - Exchange method based on children's mood and device - Google Patents

Exchange method based on children's mood and device Download PDF

Info

Publication number
CN108536802A
CN108536802A CN201810290987.3A CN201810290987A CN108536802A CN 108536802 A CN108536802 A CN 108536802A CN 201810290987 A CN201810290987 A CN 201810290987A CN 108536802 A CN108536802 A CN 108536802A
Authority
CN
China
Prior art keywords
emotional characteristics
children
mood
child user
word
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810290987.3A
Other languages
Chinese (zh)
Other versions
CN108536802B (en
Inventor
黄鸣夏
钱隽夫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Baidu Netcom Science and Technology Co Ltd
Shanghai Xiaodu Technology Co Ltd
Original Assignee
Beijing Baidu Netcom Science and Technology Co Ltd
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 Beijing Baidu Netcom Science and Technology Co Ltd filed Critical Beijing Baidu Netcom Science and Technology Co Ltd
Priority to CN201810290987.3A priority Critical patent/CN108536802B/en
Publication of CN108536802A publication Critical patent/CN108536802A/en
Application granted granted Critical
Publication of CN108536802B publication Critical patent/CN108536802B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/279Recognition of textual entities
    • G06F40/284Lexical analysis, e.g. tokenisation or collocates

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Machine Translation (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

The present invention provides a kind of exchange method, device, electronic equipment and readable storage medium storing program for executing based on children's mood, by the interactive voice information for receiving child user;Interaction content is determined in the interactive voice information;And the phonetic feature of child user is determined in the interactive voice information;Then, according to interaction content, the emotional characteristics of child user are determined;According to phonetic feature, the degree value of the emotional characteristics is determined;According to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.It realizes accurate grasp children's mood, promotes the concordance and smoothness exchanged with children, ensure that the benign guiding to children's mood.

Description

Exchange method based on children's mood and device
Technical field
The present invention relates to Emotion identification technology more particularly to a kind of exchange methods and device based on children's mood, and Using the electronic equipment and readable storage medium storing program for executing of this method.
Background technology
With the rise of artificial intelligence, human-computer interaction device has been applied in daily life more and more widely.
By taking children-story machine as an example, the man-machine interaction mode that children-story machine in the prior art generally uses is according to youngster Literal intention in children's stories language returns corresponding as a result, carrying out the interaction of " question-response ", for example, Story machine inquiry children " think Any story listened", children answer " I wants to listen small red cap ", then Story machine is found and keyword " small red cap " phase in the database Corresponding story resource simultaneously plays for it.
Man-machine interaction mode between this children and Story machine, on the one hand, since the ability to express of children is limited, not necessarily The story title wanted to listen to can accurately be said so that Story machine is that it randomly chooses story and plays, but the story and differs Surely it is that children want to listen to;On the other hand, children generally prefer that the Story machine that will talk with is considered as friend and shares with it The mood of oneself cannot give suitable however, Story machine is all answered with a kind of relatively-stationary language according to the mood of children Suitable answer, and be also possible that because it is asked it is non-answered, and the case where influence children's mood.
Therefore, the interactive mode of existing children-story machine compares mechanization and stiff, cannot be well understood by children, no The ability to exchange culture of the children of phase is formed conducive to language.
Invention content
To solve problems of the prior art, the present invention provide a kind of exchange method based on children's mood, device, Electronic equipment and readable storage medium storing program for executing, by carrying out the analysis of content and phonetic feature to the interactive voice information of child user, It accurately determines its mood, and corresponding acknowledgment strategy is provided according to its mood.Realize accurate grasp children's mood, promotion and children The concordance and smoothness of exchange ensure that the benign guiding to children's mood.
In a first aspect, the embodiment of the present invention provides a kind of exchange method based on children's mood, including:
Receive the interactive voice information of child user;
Interaction content is determined in the interactive voice information;
The phonetic feature of child user is determined in the interactive voice information;
According to the interaction content, the emotional characteristics of the child user are determined;
According to the phonetic feature, the degree value of the emotional characteristics is determined;
According to the emotional characteristics, the degree value of the emotional characteristics determines acknowledgment strategy.
Optionally, the interaction content determining in the interactive voice information, including:
The interactive voice information is converted into text message;
Semantic analysis is carried out to the text message, Topic word and/or emotion word are extracted in the text message, Obtain interaction content;
Correspondingly, described determine the emotional characteristics of the child user according to the interaction content, including:
The feelings of the child user are determined according to the Topic word and/or emotion word based on preset recognition rule Thread feature.
Optionally, further include:
Based on childrenese, children's dictionary is established;
Wherein, the childrenese includes at least:The folded word representation language of children, animal are personalized representation language;It is described Children's dictionary includes:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese;
The mood mark of the childrenese is for identifying positive mood classification, negative emotions classification, neutral mood classification.
Optionally, described to be based on preset recognition rule, according to the Topic word and/or emotion word, determine described in The emotional characteristics of child user, including:
Each Topic word and/or the corresponding mood mark of each emotion word are determined in children's dictionary Know;
If it is determined that each mood mark in do not include negative emotions classification, and include positive mood classification, it is determined that The emotional characteristics of the child user are front emotional characteristics;
If it is determined that each mood mark in do not include positive mood classification, and include negative emotions classification, it is determined that The emotional characteristics of the child user are negative emotions feature;
If it is determined that each mood mark in only include neutral mood classification, it is determined that the mood of the child user is special Sign is neutral emotional characteristics;
If it is determined that each mood mark in not only included positive mood classification, but also include negative emotions classification, then basis The word order of the Topic word and/or emotion word in the interactive voice information is based on the context meaning of one's words, determines the youngster The emotional characteristics of virgin user.
Optionally, further include:
Children's mood labeled data is obtained, training obtains children's Emotion identification model;
Correspondingly, described be based on preset recognition rule, according to the Topic word and/or emotion word, determine described in The emotional characteristics of child user, including:
The Topic word and/or emotion word are input in children's Emotion identification model, identification obtains described The emotional characteristics of child user.
Optionally, the phonetic feature that child user is determined in the interactive voice information, including:
At least one in following phonetic feature, voice intensity, word speed, intonation are determined in the interactive voice information.
Optionally, according to the phonetic feature, the degree value of the emotional characteristics is determined, including:
Using the interactive voice information integrally as objects of statistics, the corresponding voice of the interactive voice information is determined The average value of feature;
According to the average value of the phonetic feature, the degree value of the emotional characteristics is determined.
Optionally, described that the degree value of the emotional characteristics is determined according to the phonetic feature, including:
Determine that the voice of each Topic word and/or the emotion word is special in the interactive voice information Sign;
According to the weighted value of different parts of speech, the phonetic feature of the interactive voice information is weighted, institute is obtained The weighted average of the corresponding phonetic feature of predicate sound interactive information;
According to the weighted average of the phonetic feature, the degree value of the emotional characteristics is determined.
Optionally, the acknowledgment strategy includes:Dialogue heart-to-heart talk response mode and/or audio resource broadcast mode;Described According to the emotional characteristics, the degree value of the emotional characteristics determines acknowledgment strategy, including:
If the emotional characteristics are negative emotional characteristics, and the degree value of the emotional characteristics is more than predetermined threshold value, really It is dialogue heart-to-heart talk response mode to determine acknowledgment strategy;Or determine acknowledgment strategy be first with the dialogue heartly talk response mode into Row response, then response is carried out with the audio resource broadcast mode.
Optionally, further include:
Determine user's portrait of the child user;Wherein, user's portrait includes at least one of following characteristics, The attribute information of child user, the history intersection record of child user, child user habitual term, child user work and rest Incidence relation between rule, the audio resource of child user hobby, geographical location and child user;
It is drawn a portrait according to the user of the determining child user, the acknowledgment strategy is optimized.
Optionally, further include:
Obtain the temporal information and/or location information for receiving the interactive voice information;
According to the temporal information and/or location information, is drawn a portrait based on the user, determine the current institute of the child user Locate scene;
It is presently in scene according to described, the acknowledgment strategy is optimized.
Optionally, further include:
According to predetermined period, the mood analysis report of the child user is generated.
Second aspect, the embodiment of the present invention provide a kind of interactive device based on children's mood, including:
Receiving module, the interactive voice information for receiving child user;
Determining module, for determining interaction content in the interactive voice information;In the interactive voice information really Determine the phonetic feature of child user;According to the interaction content, the emotional characteristics of the child user are determined;According to the voice Feature determines the degree value of the emotional characteristics;According to the emotional characteristics, the degree value of the emotional characteristics determines response Strategy.
Optionally, the determining module, including:
Transform subblock, for the interactive voice information to be converted to text message;
Submodule is analyzed, for carrying out semantic analysis to the text message;
Extracting sub-module, for extracting Topic word and/or emotion word in the text message, interacted in Hold;
Emotional characteristics determination sub-module, for being based on preset recognition rule, according to the Topic word and/or emotion word Language determines the emotional characteristics of the child user.
Optionally, further include:
Lexicon module establishes children's dictionary for being based on childrenese;
Wherein, the childrenese includes at least:The folded word representation language of children, animal are personalized representation language;It is described Children's dictionary includes:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese;
The mood mark of the childrenese is for identifying positive mood classification, negative emotions classification, neutral mood classification.
Optionally, the emotional characteristics determination sub-module is specifically used for determining each master in children's dictionary Write inscription language and/or the corresponding mood mark of each emotion word;
When not including negative emotions classification in each mood mark determined, and including positive mood classification, determine The emotional characteristics of the child user are front emotional characteristics;
When not including positive mood classification in each mood mark determined, and including negative emotions classification, determine The emotional characteristics of the child user are negative emotions feature;
When only including neutral mood classification in each mood mark determined, determine that the mood of the child user is special Sign is neutral emotional characteristics;
When not only having included positive mood classification in each mood mark determined, but also when including negative emotions classification, according to The word order of the Topic word and/or emotion word in the interactive voice information is based on the context meaning of one's words, determines the youngster The emotional characteristics of virgin user.
Optionally, further include:
Identification model module, for obtaining children's mood labeled data, training obtains children's Emotion identification model;
Correspondingly, the emotional characteristics determination sub-module, is specifically used for inputting the Topic word and/or emotion word Into children's Emotion identification model of the identification model module, identification obtains the emotional characteristics of the child user.
Optionally, the determining module, including:
Phonetic feature determination sub-module, for determining at least one in following phonetic feature in the interactive voice information , voice intensity, word speed, intonation.
Optionally, the determining module, including:
First degree value determination sub-module, described in, integrally as objects of statistics, being determined using the interactive voice information The average value of the corresponding phonetic feature of interactive voice information;According to the average value of the phonetic feature, the mood is determined The degree value of feature.
Optionally, the determining module, including:
Second degree value determination sub-module, for determine in the interactive voice information each Topic word and/or The phonetic feature of the emotion word;According to the weighted value of different parts of speech, to the phonetic feature of the interactive voice information It is weighted, obtains the weighted average of the corresponding phonetic feature of the interactive voice information;According to the voice The weighted average of feature determines the degree value of the emotional characteristics.
Optionally, the acknowledgment strategy includes:Dialogue heart-to-heart talk response mode and/or audio resource broadcast mode;It is described true Cover half block, including:
First determination sub-module, for working as the emotional characteristics for negative emotional characteristics, and the journey of the emotional characteristics When angle value is more than predetermined threshold value, determine that acknowledgment strategy is dialogue heart-to-heart talk response mode;Or determine acknowledgment strategy be first with The dialogue heart-to-heart talk response mode carries out response, then carries out response with the audio resource broadcast mode.
Optionally, the determining module further includes:
User's portrait determination sub-module, for determining that the user of the child user draws a portrait;Wherein, the user, which draws a portrait, wraps Include at least one of following characteristics, the attribute information of child user, the history intersection record of child user, child user habit Pass between inertia term, the work and rest rule of child user, the audio resource of child user hobby, geographical location and child user Connection relationship;
Optimize submodule, the user of the child user for being determined according to user portrait determination sub-module draws Picture optimizes the acknowledgment strategy.
Optionally, described device further includes:
Acquisition module, for obtaining the temporal information and/or location information for receiving the interactive voice information;
The determining module further includes:
Scene determination sub-module, for according to the temporal information and/or location information, being drawn a portrait based on the user, really The fixed child user is presently in scene;
The optimization submodule is additionally operable to be presently in scene according to, be optimized to the acknowledgment strategy.
Optionally, further include:
Generation module, for according to predetermined period, generating the mood analysis report of the child user.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, including:
Processor;Memory;And program;Wherein, described program is stored in the memory, and is configured as It is executed by the processor, described program includes the instruction for executing method as described in relation to the first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of electronic equipment readable storage medium storing program for executing, and the electronic device-readable is deposited Storage media has program stored therein, and described program makes electronic equipment execute the method described in first aspect.
Exchange method, device, electronic equipment and readable storage medium storing program for executing provided by the invention based on children's mood, by connecing Receive the interactive voice information of child user;Interaction content is determined in the interactive voice information;And in the interactive voice information Determine the phonetic feature of child user;Then, according to interaction content, the emotional characteristics of child user are determined;According to phonetic feature, Determine the degree value of the emotional characteristics;According to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.Realize the accurate palm Children's mood is held, the concordance and smoothness exchanged with children is promoted, ensure that the benign guiding to children's mood.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention without having to pay creative labor, may be used also for those of ordinary skill in the art With obtain other attached drawings according to these attached drawings.
Fig. 1 be an exemplary embodiment shown in the present invention is based on the flow charts of the exchange method of children's mood;
Fig. 2 be another exemplary embodiment shown in the present invention is based on the flow charts of the exchange method of children's mood;
Fig. 3 be an exemplary embodiment shown in the present invention is based on the structural schematic diagrams of the interactive device of children's mood;
Fig. 4 be another exemplary embodiment shown in the present invention is based on the structural schematic diagrams of the interactive device of children's mood;
Fig. 5 a are the structural schematic diagram of the electronic equipment of the present invention shown in an exemplary embodiment;
Fig. 5 b are the structural schematic diagram of the electronic equipment of the present invention shown in another exemplary embodiment.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
Term " first ", " second ", " third " in description and claims of this specification and above-mentioned attached drawing and " The (if present)s such as four " are for distinguishing similar object, without being used to describe specific sequence or precedence.It should manage The data that solution uses in this way can be interchanged in the appropriate case, so as to the embodiment of the present invention described herein can in addition to Here the sequence other than those of diagram or description is implemented.In addition, term " comprising " and " having " and their any deformation, Be intended to cover it is non-exclusive include, for example, containing the process of series of steps or unit, method, system, product or setting It is standby those of to be not necessarily limited to clearly to list step or unit, but may include not listing clearly or for these mistakes The intrinsic other steps of journey, method, product or equipment or unit.
It should be noted that technical scheme of the present invention is described in detail with specific embodiment below.Below this Several specific embodiments can be combined with each other, may be no longer superfluous in some embodiments for same or analogous concept or process It states.
Fig. 1 be an exemplary embodiment shown in the present invention is based on the flow charts of the exchange method of children's mood, such as Fig. 1 institutes Show, an embodiment of the present invention provides a kind of exchange method based on children's mood, this method can be by arbitrarily executing based on children The device of the exchange method of mood executes, which can pass through software and or hardware realization.In the present embodiment, executing should The device of exchange method based on children's mood can be integrated in the electronic equipment of user, can also be integrated in the electricity with user Sub- equipment carries out in the cloud server of network data exchange.Wherein, electronic equipment, including but not limited to children's interactive device (Story machine, smartwatch, interaction robot), smart mobile phone, tablet computer, portable computer and desktop computer etc..Its By receiving the interactive voice information of children, the exchange method based on children's mood based on the present invention is to the interactive voice information Analysis and identification is carried out, the acknowledgment strategy for meeting child user current emotional is obtained, in a manner of more understanding children's mood It is smoothly exchanged with children.Cloud server may include one or more for executing the functions such as operation, analysis, storage Server, by execute the exchange method based on children's mood device be Story machine for, Story machine receive children language Sound interactive information is sent to cloud server, after cloud server receives the interactive voice information, based on the present invention based on The exchange method of children's mood carries out analysis and identification to the interactive voice information, obtains the response for meeting child user current emotional The acknowledgment strategy is fed back to Story machine by network transmission, realizes Story machine in a manner of more understanding children's mood by strategy It is smoothly exchanged with children.The executive agent in each embodiment is illustrated by taking Story machine as an example below, this implementation Example the exchange method based on children's mood include:
Step 101, the interactive voice information for receiving child user.
In the present embodiment, the audio receiver part in Story machine, such as microphone, can receive the voice of child user Interactive information, the voice messaging which is sent out by the children that Story machine senses.Wherein, interactive voice information Can include the instruction language that children send out to Story machine, for example, " I wants to listen the event of Sofia princess instruction Story machine Thing ", " we two chat " etc.;Can also include only the acoustic information that children send out, for example, crying sound, laugh, or It is sigh " eh~~", ", eh~~~" etc..
Step 102 determines interaction content in interactive voice information.
In the present embodiment, so-called interaction content, the intention exactly identified in interactive voice information, for example, children When user expresses it and wants to listen the intention of the story of Sofia princess, can may not accurately it be given expression to as adult Come, but during expressing its wish, it adds pause, elongate the various disturbing factors such as sound, folded word, for example, children use The interactive voice information that family gives expression to may be, " I ... uh ... Sofia ... mews~~~~mew~~~~~ .. " is heard ..., then it can be with text that Story machine, which needs the interaction content from being determined in the interactive voice information, the interaction content, The content that font formula can give expression to, for example, the interaction content that Story machine is determined can be " I hears uh Sofia mew mew "; Story machine can also be according to preset recognition rule, for example, removal interjection, folds word simplification, the rules such as word order adjustment, in the language The interaction content of " I listens Sofia, mew " is determined in sound interactive information.So that Story machine can be according to from interactive voice Interaction content is determined in information, and its meaning of one's words is identified according to interaction content, to determine the intention of children.For in interactive voice What obtains interaction content from information, the technologies such as semantics recognition, word segmentation processing in the prior art may be used, the present embodiment is to this It is not especially limited.
Step 103, the phonetic feature that child user is determined in interactive voice information.
In the present embodiment, so-called phonetic feature is the feature for being assessed sound and being described.It may include, Reflect the sound intensity of acoustic energy;Reflect the loudness of sound intensity degree;For example, one timing of sound frequency, the sound intensity is stronger, and loudness is got over Greatly;Can also include that the pitch that reflection auditory system feels sound frequency height, the fundamental tone of sounder gender is all for identification Phase, the features such as signal-to-noise ratio.Wherein, characteristic dimension included in phonetic feature can be by those skilled in the art according to identification need Ask voluntarily determining, the present embodiment is not especially limited this.By analyzing phonetic feature, the attribute of child user is aided in determining whether Information, for example, child user is a Boy or a girl;The sound ray feature reflected according to phonetic feature can also determine children's Substantially age level etc.;To provide prediction data for description user's portrait.In addition, in short carrying out table with different loudness of a sound Up to when, phonetic feature can also reflect the emotional state of speaker, for example, children expression " I wants to listen the event of Sofia princess It is expressed with more soft intonation when thing ", then can determine that children's mood is more gentle at this time, predict that it is really intended to listen This story;And when identical content of the children with the intonation expression " I wants to listen the story of Sofia princess " shouted of neighing, it is likely that Children just gave birth to quarrel with human hair at this time, very indignant, were really that situation can although then children express it and want to listen the story Can be that children not really want to listen story, but using Story machine as object is given vent to, to its release mood, then this moment if Children play the story and seem just not instead of very unsuitable, should according to the mood of children this moment, using other exchange ways, With the mood for its indignation of releiving.It therefore, can be more accurately by determining the phonetic feature of children from interactive voice information Grasp the emotional state of children.
Step 104, according to interaction content, determine the emotional characteristics of child user.
In the present embodiment, speech is usually all expression mood most direct mode, and mood is divided with major class, can be general It includes as pleasure, anger, sorrow, happiness;By the identification to word, the mood of speaker can be relatively accurately grasped.For example, children User may say " today is very happy " in the emotion of expression " happiness ";The then interaction content by being determined in step 102 " happy " can primarily determine that the emotional characteristics of children tend to happily;And when emotion of the children in expression " anger ", it may It says " I dislikes you ";The emotional characteristics of children can be then primarily determined by the interaction content " disagreeable " determined in step 102 Tend to unhappy or angry.Wherein, for the classification of emotional characteristics, it can be derived based on pleasure, anger, sorrow, happiness and more be refined Classification, it is glad, like for example, happiness is further divided into;Anger is further divided into, angry, hatred;In addition, it can include mood Feature has, frightened, fears.By the way that emotional characteristics are carried out with the subdivision of classification or subclass, the accuracy rate of Emotion identification is promoted.
Step 105, according to phonetic feature, determine the degree value of emotional characteristics.
In the present embodiment, as described in preceding step 103, the features such as pitch, tone, word speed in phonetic feature can be one Determine the mood for reflecting speaker in degree, for example, example a, when mother softly says child " quickly eat eh~~" then Reflect on the time to be not very urgent to a certain extent, may only desire to child can more attentively be at table this part thing A kind of communicative habits in feelings;If however, example b, mother neigh to child shout as say it is " fast!Point!It eats!!!!", it is likely that The case where reflecting is that the mood of mother is anger;It is really very urgent to be also possible to the time.Therefore, it is based on phonetic feature Can be to the emotional characteristics that step 104 is determined, such as the emotional characteristics of " worried " carry out the assessment of degree value, for example, for upper The case where stating example a, the mood degree value of " worried " are less than the mood degree value in example b;That is example a has reflected " one As worry ", the emotional characteristics of " time is not at full stretch ";Example b has reflected the mood of " worrying very much ", " very nervous " Feature.
Step 106, according to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.
In the present embodiment, it according to the emotional characteristics determined from interaction content, and is determined according to from phonetic feature The degree value of the emotional characteristics gone out determines acknowledgment strategy corresponding with interaction content and meeting children's mood.For example, it receives To child user interactive voice information be children cried and said " I am unhappy today, toot~~~~", therefrom determine to hand over Mutual content is " I am unhappy today, toot ";Then according to the interaction content, determine that the emotional characteristics of children are " unhappy ", " hinder The moods such as the heart ";Based on the crying that children express in the interactive information, determine that the phonetic feature of children is that can be identified for that out The number voice feature (e.g., the intensity of sound of sobs, the features such as duration for sound of sobbing) of " sobs ", then according to should The degree value of the phonetic feature of " sobs ", the emotional characteristics " unhappy ", " sad " that determine children is " moderate is sad ", or " unhappy in the extreme ".It to which the acknowledgment strategy that Story machine may be used can be, heartly talks with children, " you how by inquiry children Why cry", or its favorite story is played for children, alleviate its sad mood to suit the medicine to the illness.
Exchange method provided in an embodiment of the present invention based on user emotion, the interactive voice by receiving child user are believed Breath;Interaction content is determined in the interactive voice information;And the phonetic feature of child user is determined in the interactive voice information; Then, according to interaction content, the emotional characteristics of child user are determined;According to phonetic feature, the degree value of the emotional characteristics is determined; According to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.It realizes accurate grasp children's mood, is promoted and handed over children The concordance and smoothness of stream ensure that the benign guiding to children's mood.
Fig. 2 be another exemplary embodiment shown in the present invention is based on the flow charts of the exchange method of user emotion.In Fig. 1 On the basis of illustrated embodiment, the exchange method based on user emotion of the present embodiment specifically includes:
Step 201, the interactive voice information for receiving child user.
Interactive voice information is converted to text message by step 202.
In the present embodiment, it may be used the Word Input with semantic information in interactive voice information in the prior art Technology out forms text message.The place occupied in the memory space and calculating process that are generally occupied due to text message It is all smaller to manage device resource, therefore, interactive voice information is translated as corresponding text and is conducive to be promoted in mood determination process The accuracy of mood analysis is advantageously ensured that the promotion of the treatment effeciency of text message, and treatment effeciency.For example, it receives Interactive voice information to children be " my, the kindergartens Qu Liao breathe out today~~~breathe out~~~~breathe out~~~breathe out~~ ~~", then transformed text message can be " my kindergartens Qu Liao today are breathed out ".Wherein, lasting laugh " is breathed out ~~~breathe out~~~~breathe out~~~breathe out~~~~" conversion, to retain the language elements that can be accurately identified as principle Text determination is carried out, for example, text message of one or more " Kazakhstan " word recognized as the laugh can be understood in laugh, Then retain " Kazakhstan " word of corresponding number in text message.
Step 203 carries out semantic analysis to text message, and Topic word and/or emotion word are extracted in text message, Obtain interaction content.
In the present embodiment, it by carrying out semantic analysis to text message, is extracted from the text message being converted to Element with break-up value, such as Topic word, emotion word;Wherein, semantic analysis may be used in the prior art Participle technique, such as at least one of following granularity:Individual Chinese character, single character, word, phrase to text message into Row participle, and according to the part of speech or the features such as grammer in participle, the Topic word that can express a word subject content is extracted, And/or the emotion word of speaker's Sentiment orientation can be expressed.Such as the example in step 202, children say that " I goes today Breathe out kindergarten " wherein, the Topic word extracted can be " kindergarten ", and emotion word can be " ", " ", " Kazakhstan " etc. Tone word.Content related with kindergarten then can be responded in acknowledgment strategy by Topic word " kindergarten ", pass through feelings Feel the tone words such as word " ", " ", " Kazakhstan ", it can be determined that the present mood of children is more happy, in acknowledgment strategy Middle cooperation is with cheerful and light-hearted content.
Step 204 determines the feelings of child user based on preset recognition rule according to Topic word and/or emotion word Thread feature.
In the present embodiment, which can be to search Topic word and/or emotion based on children's dictionary Word corresponding emotional characteristics in dictionary.Wherein, so-called children's dictionary, is to be established based on childrenese, that is to say, that youngster Virgin dictionary is the dictionary formed for the speech habits in childrenese.
Wherein, childrenese at least may include:The folded word representation language of children, animal are personalized representation language etc.;It should Can include in children's dictionary:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese Deng;Wherein, the mood mark of childrenese can be used for identifying positive mood classification, negative feelings in the language expressed by children Thread classification or neutral mood classification.
For example, children like folded word, such as " carefully ", " eating fruit fruit ", then in addition to there is the paraphrase of " good " in children's dictionary, There can also be the paraphrase to folding word " carefully ", can be such as " unusual good, very like " to the paraphrase for folding word " carefully ";Institute Call childrenese combination context paraphrase, be based on childrenese expression it is typically no be grown up it is rigorous, can in communicative habits It can will appear that pass judgement on justice mixed, situations such as front and back word order is chaotic, therefore, look like only according to the word face obtained after word segmentation processing true The meaning made may be not the original idea of children, therefore, can be based on context and carry out paraphrase to word, for example, children say " it is husky that mother sits hair ", its paraphrase may and be not present in word " hair is husky " in children's dictionary, but combines context according to " hair It is husky " before " seat " may determine that the word order that " hair is husky " be " sofa " is reverse, then combine context can correctly to " hair sand " into Row paraphrase.For another example children say " I does not like well you, relative's Pussy ", it is not like according to " no " determining that children to be expressed Vigorously, still, based on context, " do not like well ", be then expressed as liking, if there is " happiness to " not liking well " in children's dictionary Vigorously " and a variety of paraphrase of " not liking ", then it can also continue to based on context " relative's Pussy ", determine " not liking well " at this Sentence in give expression to be " liking ", " happiness " mood.It therefore, can be in children's dictionary for " not liking well " this word The moods such as " liking ", " happiness ", " happy " mark is identified.Mood mark indicates that the present mood of children is positive feelings Thread classification, and moods classification, the those skilled in the art such as front, negative, neutrality can be finely divided according to statistical data Class promotes the accuracy of Emotion identification to realize more horn of plenty and accurate children's dictionary.
Based on children's dictionary, it can determine that each Topic word and/or each emotion word correspond in children's dictionary Mood mark;
A) if it is determined that each mood mark in do not include negative emotions classification, and include positive mood classification, then really The emotional characteristics for determining child user are front emotional characteristics;
Specifically, Topic word and emotion word that interactive voice information extraction goes out may have it is multiple, then by children It is inquired in dictionary, with the emotional characteristics of each word of determination.If it is determined that the emotional characteristics that each word gives expression to are positive, Either positive or neutral, then it can determine that the emotion of whole word expression is positive mood.For example, children say " cat today Good beauty ", then the Topic word extracted can be " cat ", " today ";Emotion word can be " good ", " beauty ";It is based on Children's dictionary " cat ", " today " mood mark can be neutral emotional characteristics, and identify can be with for the mood of " good ", " beauty " For front emotional characteristics, accordingly, it is determined that the emotional characteristics of whole word expression are front emotional characteristics.
B) if it is determined that each mood mark in do not include positive mood classification, and include negative emotions classification, then really The emotional characteristics for determining child user are negative emotions feature;
For example, children say " I dislikes cat ", then the Topic word extracted can be " cat ", wherein " I " base Topic word is can be determined that in the rule of Topic word extraction, can also be confirmed as personal pronoun and is ignored, ability Field technique personnel can be determined according to preset semantic analysis rule, and the present embodiment is not especially limited this.In the sentence Emotion word be " disagreeable ";Mood mark based on children's dictionary " cat ", " I " can be neutral emotional characteristics, and " beg for Detest " mood mark can be negative emotions feature, accordingly, it is determined that whole word expression emotional characteristics be negative emotions feature.
C) if it is determined that each mood mark in only include neutral mood classification, it is determined that the emotional characteristics of child user For neutral emotional characteristics;
For example, children say " my kindergartens Qu Liao today ", then the Topic word extracted can be " today ", " children Youngster garden ", or " I ", " today ", " going ", " kindergarten " do not include emotion word in the sentence, then are based on children's word Allusion quotation, these above-mentioned words can be identified as neutral emotional characteristics, it is determined that the emotional characteristics of whole word expression are neutral mood Feature.
D) if it is determined that each mood mark in not only included positive mood classification, but also include negative emotions classification, then root According to the word order of Topic word and/or emotion word in interactive voice information, it is based on the context meaning of one's words, determines the feelings of child user Thread feature.
For example, children say " I well dislike cat ", then the Topic word extracted can be " cat ", in this Emotion word can be " good ", " disagreeable ";Mood mark based on children's dictionary " cat " can be neutral emotional characteristics, and The mood mark of " good " can be front emotional characteristics, and the mood mark of " disagreeable " can be negative emotions feature, then at this time may be used With the word order according to " good " and " disagreeable " in former sentence, it is based on the context meaning of one's words, determines that " good disagreeable " is negative emotions feature.It needs , both can be according to D it is noted that for the Emotion identification of " good disagreeable ") in identification method be determined as negative emotions, also may be used Can be according to the communicative habits of children, it is " good to be determined in the paraphrase of childrenese combination context included in children's dictionary It is disagreeable " emotional characteristics as entire term identification " not liking ".
The preset recognition rule searches Topic word and/or emotion word in dictionary in addition to that can be based on children's dictionary In corresponding emotional characteristics, be also based on children's Emotion identification model, to childrenese express in mood be identified.
Firstly, it is necessary to establish children's Emotion identification model, by obtaining children's mood labeled data, training obtains children Emotion identification model;Then Topic word and/or emotion word are input in children's Emotion identification model, identification obtains The emotional characteristics of virgin user.
It is to be based on a large amount of statistical data, obtained lexicographic rule searching for the mode of former children's dictionary; And for the ever-changing of language, and the differentiation of the popular language occurred at any time is followed, the update of children's dictionary will be by Therefore larger challenge can be based on the advanced algorithms such as neural network, be trained to children's idiom of big data quantity, So that the model is passed through the tongue for constantly learning children, obtains that there is recognition capability to the mood gone out expressed by childrenese Children's Emotion identification model accurately identifies children's mood with realizing, and can be by constantly learning to carry out more with training Newly, the accuracy rate of identification is promoted.
Step 205, determined in interactive voice information in following phonetic feature at least one of, voice intensity, word speed, language It adjusts.
In the present embodiment, voice intensity can include the sound characteristics such as the sound intensity, loudness, pitch;
Word speed can reflect that children's is nervous, urgent, cheerful and light-hearted, excited by the size of vocabulary included in the unit interval Equal Emotion expressions;
Intonation is the intonation spoken, and instruction height rises and falls the preparation and variation of weight.The intonation meaning of a word can be with table The attitude or implication of speaker are shown.Same sentence, intonation is different, and the meaning will be different.For example, children say that " I wants today Do one's assignment " according to intonation be stable intonation or more droning intonation, may indicate that it is not very willing to do one's assignment, less Happy mood;And according to intonation be rising tune, then the word given expression to question sentence " I will do one's assignment today "Effect.
Therefore, by the identification to voice intensity, word speed, intonation in interactive voice information etc., child user is specified Mood is inclined to.
Step 206, according to phonetic feature, determine the degree value of emotional characteristics.
In the present embodiment, when the degree value for determining emotional characteristics, can the phonetic feature based on whole word into Row statistics, can also be counted based on the phonetic feature of each word in whole word.Specifically,
First way determines the corresponding voice of interactive voice information using interactive voice information integrally as objects of statistics The average value of feature;According to the average value of phonetic feature, the degree value of emotional characteristics is determined.
The second way determines the phonetic feature of each Topic word and/or emotion word in interactive voice information;According to The phonetic feature of interactive voice information is weighted in the weighted value of different parts of speech, and it is corresponding to obtain interactive voice information The weighted average of phonetic feature;According to the weighted average of phonetic feature, the degree value of emotional characteristics is determined.
Above two mode cuts both ways, first way, according to the phonetic feature that whole sentence is expressed, such as according to speech wave Shape, statistics obtain the average value of phonetic feature, which can be compared with preset threshold value, to determine that it is reflected Emotional characteristics degree value.Its benefit is that algorithm amount is small, and algorithm is simple, effectively promotes the degree value for determining emotional characteristics Efficiency.But being averaged of carrying out of the expression by being then based on whole sentence, sometimes in whole sentence may there is only the intonation of individual words or Voice intensity has than stronger variation, then when carrying out taking average operation to whole phonetic feature, the effect of this reinforcing has It may be weakened, average out.The second way, to the voice of each Topic word and/or emotion word in whole sentence Feature extracts, and imparts different weighted values also according to the part of speech of word, by being weighted processing to each word, And average computation is weighted to whole sentence, obtain degree value.The degree value that this mode is determined is more accurate, but due to being The phonetic feature of each word is determined, and is weighted, therefore, compared to first way, calculation amount has It is a little big, but accuracy rate can be promoted.For example, child user say " I not!!!Like eating radish ", in expression word High tone, and the part of speech of " no " as negative word are used when " no ", are endowed larger weighted value, then weighted average calculation Afterwards, the degree value of the negative emotions gone out expressed by this can be enhanced.
Step 207, according to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.
In the present embodiment, acknowledgment strategy may include:Dialogue heart-to-heart talk response mode and/or audio resource broadcast mode; Then, if emotional characteristics are negative emotional characteristics, and the degree value of emotional characteristics is more than predetermined threshold value, it may be determined that acknowledgment strategy For dialogue heart-to-heart talk response mode;Or determine that acknowledgment strategy is first to carry out response to talk with heart-to-heart talk response mode, then provide with audio Source broadcast mode carries out response.
That is, when children be in mood it is very low when, it should the response in the form of more deeply concerned, such as and its It engages in the dialogue, to understand its depressed reason;If at this point, directly responding children with music or story, children can be made to feel Feel that the mood of oneself is ignored, and it is increasingly depressed.Therefore, for negative emotional characteristics, acknowledgment strategy can be Talk with the response mode of heart-to-heart talk;Or first talk with heart-to-heart talk with it, further according to its emotional change, intert into audio resource (such as children The song or story liked) carry out mood releive.
Step 208, according to the user of determining child user draw a portrait and/or according to scene is presently in, to acknowledgment strategy It optimizes.
In the present embodiment, when determining acknowledgment strategy, in addition to according in 201~step 207 of above-mentioned steps to youngster Virgin interactive voice information is analyzed so that it is determined that outside acknowledgment strategy, which can be combined with user's portrait of children (what e.g., a Boy or a girl likes, and dislikes what etc.) is determined.The method of determination of relevant user portrait, this field skill The method for determining user's portrait in the prior art may be used in art personnel, such as analyzes the previous intersection record of user and is determined. The present embodiment does not limit this specifically.
Wherein, user's portrait may include at least one of following characteristics, attribute information, the child user of child user History intersection record, the habitual term of child user, the work and rest rule of child user, child user hobby audio money Incidence relation etc. between source, geographical location and child user;To be drawn a portrait according to the user of determining child user, to response Strategy optimizes.
In addition, Story machine can also obtain the temporal information and/or location information for receiving interactive voice information;According to the time Information and/or location information are drawn a portrait based on user, determine that child user is presently in scene;According to scene is presently in, correspond to Strategy is answered to optimize.
For example, according to 201~step 207 of abovementioned steps, determine that the current mood of children is happy, however, this When the location information that gets be the children family, temporal information is 9 points at night, then is drawn a portrait, found according to the user of children The general sack time of children be 9 points 30 minutes or so, then according to step 207 determine acknowledgment strategy may be play one section it is cheerful and light-hearted Music, it is contemplated that current scene be children fall asleep before, then can select in cheerful and light-hearted music one head more releive Song, to contribute to children punctual to fall asleep within 30 minutes or so at 9 points.Therefore, (such as user draws with factor the considerations of various dimensions Picture, time, place etc.) it is combined, can acknowledgment strategy more be optimized, make it in the same of the mood for meeting child user When, help to guide its mood to more benign direction.
Optionally, the mood analysis report that child user according to predetermined period, can also be generated, so that parent can be more Understand the child of oneself.
Fig. 3, should for the present invention is based on the structural schematic diagrams of the interactive device of children's mood shown in an exemplary embodiment Device can be the various electronic equipments for supporting information communication, including but not limited to children's interactive device (Story machine, intelligent hand Table, interaction robot), smart mobile phone, tablet computer, portable computer and desktop computer etc..This is based on children's mood Interactive device can also be arranged in server beyond the clouds, cloud server by with children's interactive device (Story machine, intelligent hand Table, interaction robot), smart mobile phone, tablet computer, the electronic equipments such as portable computer and desktop computer carry out network number According to interaction, the service that acknowledgment strategy is provided for children's mood is realized.Wherein, cloud server may include one or more use In the server for executing the functions such as operation, analysis, storage, by taking children's interactive device is Story machine as an example, Story machine receives children Interactive voice information, be sent to cloud server, after cloud server receives the interactive voice information, be based on aforementioned implementation The exchange method based on children's mood in example carries out analysis and identification to the interactive voice information, obtains meeting child user current The acknowledgment strategy is realized Story machine more to understand children's mood by the acknowledgment strategy of mood by network-feedback to Story machine Mode smoothly exchanged with children.The interactive device based on children's mood can pass through software, hardware or soft or hard The mode that part combines is realized.As shown in figure 3, the device includes:
Receiving module 31, the interactive voice information for receiving child user.
Determining module 32, for determining interaction content in interactive voice information;Children are determined in interactive voice information The phonetic feature of user;According to interaction content, the emotional characteristics of child user are determined;According to phonetic feature, emotional characteristics are determined Degree value;According to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.
Interactive device provided in this embodiment based on children's mood can execute above-mentioned embodiment of the method shown in FIG. 1, Its implementing principle and technical effect is similar, and details are not described herein.
Interactive device provided in this embodiment based on user emotion, by the interactive voice information for receiving child user; Interaction content is determined in the interactive voice information;And the phonetic feature of child user is determined in the interactive voice information;So Afterwards, according to interaction content, the emotional characteristics of child user are determined;According to phonetic feature, the degree value of the emotional characteristics is determined;Root According to emotional characteristics, the degree value of emotional characteristics determines acknowledgment strategy.Realize that accurate grasp children's mood, promotion are exchanged with children Concordance and smoothness, ensure that the benign guiding to children's mood.
Fig. 4 is shown in another exemplary embodiment the present invention is based on the structural schematic diagram of the interactive device of children's mood, As shown in figure 4, on the basis of the above embodiments, which further includes:
Determining module 32, including:
Transform subblock 321, for interactive voice information to be converted to text message.
Submodule 322 is analyzed, for carrying out semantic analysis to text message.
Extracting sub-module 323 obtains interaction content for extracting Topic word and/or emotion word in text message.
Emotional characteristics determination sub-module 324, for being based on preset recognition rule, according to Topic word and/or emotion word Language determines the emotional characteristics of child user.
Optionally, further include:
Lexicon module 33 establishes children's dictionary for being based on childrenese.
Wherein, childrenese includes at least:The folded word representation language of children, animal are personalized representation language;Children's dictionary Including:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese.
The mood mark of childrenese is for identifying positive mood classification, negative emotions classification, neutral mood classification.
Optionally, emotional characteristics determination sub-module 324, be specifically used for determining in children's dictionary each Topic word and/ Or each the corresponding mood of emotion word identifies.
When not including negative emotions classification in each mood mark determined, and including positive mood classification, determine The emotional characteristics of child user are front emotional characteristics;
When not including positive mood classification in each mood mark determined, and including negative emotions classification, determine The emotional characteristics of child user are negative emotions feature;
When only including neutral mood classification in each mood mark determined, determine that the emotional characteristics of child user are Neutral emotional characteristics;
When not only having included positive mood classification in each mood mark determined, but also when including negative emotions classification, according to The word order of Topic word and/or emotion word in interactive voice information is based on the context meaning of one's words, determines the mood of child user Feature.
Optionally, further include:
Identification model module 34, for obtaining children's mood labeled data, training obtains children's Emotion identification model.
Correspondingly, emotional characteristics determination sub-module 324, specifically for Topic word and/or emotion word are input to knowledge In children's Emotion identification model of other model module, identification obtains the emotional characteristics of child user.
Optionally, determining module 32, including:
Phonetic feature determination sub-module 325, for determining at least one in following phonetic feature in interactive voice information , voice intensity, word speed, intonation.
Optionally, determining module 32, including:
First degree value determination sub-module 326, for, integrally as objects of statistics, determining that voice is handed over using interactive voice information The average value of the corresponding phonetic feature of mutual information;According to the average value of phonetic feature, the degree value of emotional characteristics is determined.
Optionally, determining module 32, including:
Second degree value determination sub-module 327, for determining each Topic word and/or emotion word in interactive voice information The phonetic feature of language;According to the weighted value of different parts of speech, the phonetic feature of interactive voice information is weighted, language is obtained The weighted average of the corresponding phonetic feature of sound interactive information;According to the weighted average of phonetic feature, emotional characteristics are determined Degree value.
Optionally, acknowledgment strategy includes:Dialogue heart-to-heart talk response mode and/or audio resource broadcast mode;Determining module 32, including:
First determination sub-module 328, for being negative emotional characteristics when emotional characteristics, and the degree value of emotional characteristics is super When crossing predetermined threshold value, determine that acknowledgment strategy is dialogue heart-to-heart talk response mode;Or determine that acknowledgment strategy is first to be answered to talk with heart-to-heart talk The mode of answering carries out response, then carries out response with audio resource broadcast mode.
Optionally, determining module 32 further include:
User's portrait determination sub-module 329, for determining that the user of child user draws a portrait;Wherein, user portrait include with At least one of lower feature, the attribute information of child user, the history intersection record of child user, child user habituation Being associated between term, the work and rest rule of child user, the audio resource of child user hobby, geographical location and child user System.
Optimize submodule 330, the user of the child user for being determined according to user's portrait determination sub-module draws a portrait, corresponding Strategy is answered to optimize.
Optionally, further include:
Acquisition module 35, for obtaining the temporal information and/or location information that receive interactive voice information.
Determining module 32 further includes:
Scene determination sub-module 331, for according to temporal information and/or location information, being drawn a portrait based on user, determining children User is presently in scene.
Optimize submodule 330, is additionally operable to, according to scene is presently in, optimize acknowledgment strategy.
Optionally, further include:
Generation module 36, for according to predetermined period, generating the mood analysis report of child user.
Interactive device provided in this embodiment based on user emotion can execute above-mentioned embodiment of the method shown in Fig. 2, Its implementing principle and technical effect is similar, and details are not described herein.
Fig. 5 a are the structural schematic diagram of the electronic equipment of the present invention shown in an exemplary embodiment.Electronic equipment 500 includes: Processing unit 502 and communication unit 503.Processing unit 502 is used to carry out control management to the action of electronic equipment 500, for example, Processing unit 502 is used to that the step 102- steps 106 that electronic equipment 500 executes Fig. 1, processing unit 502 to be supported to can be also used for Electronic equipment 500 is supported to execute the step 202- steps 208 of Fig. 2, and/or other mistakes for technology described in the invention Journey.Communication unit 503 is used for the communication of electronic equipment 500 and other network entities, can be also used for that electronic equipment 500 is supported to hold The step 101 of row Fig. 1 or the step 201 of Fig. 2.Electronic equipment 500 can also include storage unit 501, for storing electronics The program code and data of equipment 500.
Wherein, processing unit 502 can be processor or controller, such as can be CPU, general processor, number letter Number processor (DigitalSignalProcessor, DSP), application-specific integrated circuit (Application- SpecificIntegratedCircuit, ASIC), field programmable gate array (FieldProgrammableGateArray, FPGA) either other programmable logic device, transistor logic, hardware component or its arbitrary combination.It may be implemented Or it executes and combines various illustrative logic blocks, module and circuit described in the disclosure of invention.The processor Can realize the combination of computing function, such as combined comprising one or more microprocessors, the combination etc. of DSP and microprocessor Deng.Communication unit 503 can be communication interface, transceiver, transmission circuit etc., wherein communication interface is to be referred to as, and may include one A or multiple interfaces.Storage unit 501 can be memory.
When processing unit 502 is processor, communication unit 503 is communication interface, when storage unit 501 is memory, this The involved electronic equipment of invention can be electronic equipment 510 shown in Fig. 5 b.
Shown in Fig. 5 b, which includes:Processor 512, communication interface 513, memory 511.It is optional , electronic equipment 510 can also include bus 514.Wherein, communication interface 513, processor 512 and memory 511 can lead to Cross the interconnection of bus 514;Bus 514 can be Peripheral Component Interconnect standard (PeripheralComponentInterconnect, abbreviation PCI) bus or expanding the industrial standard structure (ExtendedIndustryStandardArchitecture, abbreviation EISA) bus etc..It is total that bus 514 can be divided into address Line, data/address bus, controlling bus etc..It for ease of indicating, is only indicated with a thick line in Fig. 5 b, it is not intended that only one total Line or a type of bus.
In addition, having program stored therein in memory 511, and it is configured as being executed by processor 512, which includes being used for Execute the instruction of the method described in any embodiment as above.
The embodiment of the present invention also provides a kind of electronic equipment readable storage medium storing program for executing, wherein electronic device-readable storage medium It has program stored therein, program makes server execute the exchange method based on user emotion that aforementioned any embodiment provides.Wherein, Above-mentioned readable storage medium storing program for executing can be realized by any kind of volatibility or non-volatile memory device or combination thereof, Such as static RAM (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only Memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk Or CD.
One of ordinary skill in the art will appreciate that:Realize that all or part of step of above-mentioned each method embodiment can lead to The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer read/write memory medium.The journey When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned includes:ROM, RAM, magnetic disc or The various media that can store program code such as person's CD.
Finally it should be noted that:The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent Present invention has been described in detail with reference to the aforementioned embodiments for pipe, it will be understood by those of ordinary skill in the art that:Its according to So can with technical scheme described in the above embodiments is modified, either to which part or all technical features into Row equivalent replacement;And these modifications or replacements, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution The range of scheme.

Claims (26)

1. a kind of exchange method based on children's mood, which is characterized in that including:
Receive the interactive voice information of child user;
Interaction content is determined in the interactive voice information;
The phonetic feature of child user is determined in the interactive voice information;
According to the interaction content, the emotional characteristics of the child user are determined;
According to the phonetic feature, the degree value of the emotional characteristics is determined;
According to the emotional characteristics, the degree value of the emotional characteristics determines acknowledgment strategy.
2. according to the method described in claim 1, it is characterized in that, described determine in the interactive voice information in interaction Hold, including:
The interactive voice information is converted into text message;
Semantic analysis is carried out to the text message, Topic word and/or emotion word are extracted in the text message, is obtained Interaction content;
Correspondingly, described determine the emotional characteristics of the child user according to the interaction content, including:
Based on preset recognition rule, according to the Topic word and/or emotion word, determine that the mood of the child user is special Sign.
3. according to the method described in claim 2, it is characterized in that, further including:
Based on childrenese, children's dictionary is established;
Wherein, the childrenese includes at least:The folded word representation language of children, animal are personalized representation language;The children Dictionary includes:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese;
The mood mark of the childrenese is for identifying positive mood classification, negative emotions classification, neutral mood classification.
4. according to the method described in claim 3, it is characterized in that, described be based on preset recognition rule, according to the theme Word and/or emotion word determine the emotional characteristics of the child user, including:
Each Topic word and/or the corresponding mood mark of each emotion word are determined in children's dictionary;
If it is determined that each mood mark in do not include negative emotions classification, and include positive mood classification, it is determined that described The emotional characteristics of child user are front emotional characteristics;
If it is determined that each mood mark in do not include positive mood classification, and include negative emotions classification, it is determined that described The emotional characteristics of child user are negative emotions feature;
If it is determined that each mood mark in only include neutral mood classification, it is determined that the emotional characteristics of the child user are Neutral emotional characteristics;
If it is determined that each mood mark in not only included positive mood classification, but also include negative emotions classification, then according to The word order of Topic word and/or emotion word in the interactive voice information is based on the context meaning of one's words, determines that the children use The emotional characteristics at family.
5. according to the method described in claim 2, it is characterized in that, further including:
Children's mood labeled data is obtained, training obtains children's Emotion identification model;
Correspondingly, described determine the children based on preset recognition rule according to the Topic word and/or emotion word The emotional characteristics of user, including:
The Topic word and/or emotion word are input in children's Emotion identification model, identification obtains the children The emotional characteristics of user.
6. according to the method described in claim 2, it is characterized in that, described determine child user in the interactive voice information Phonetic feature, including:
At least one in following phonetic feature, voice intensity, word speed, intonation are determined in the interactive voice information.
7. according to the method described in claim 2, it is characterized in that, according to the phonetic feature, the emotional characteristics are determined Degree value, including:
Using the interactive voice information integrally as objects of statistics, the corresponding phonetic feature of the interactive voice information is determined Average value;
According to the average value of the phonetic feature, the degree value of the emotional characteristics is determined.
8. according to the method described in claim 2, it is characterized in that, described according to the phonetic feature, determine that the mood is special The degree value of sign, including:
Determine the phonetic feature of each Topic word and/or the emotion word in the interactive voice information;
According to the weighted value of different parts of speech, the phonetic feature of the interactive voice information is weighted, obtains institute's predicate The weighted average of the corresponding phonetic feature of sound interactive information;
According to the weighted average of the phonetic feature, the degree value of the emotional characteristics is determined.
9. according to the method described in claim 1, it is characterized in that, the acknowledgment strategy includes:Dialogue heart-to-heart talk response mode, And/or audio resource broadcast mode;It is described that according to the emotional characteristics, the degree value of the emotional characteristics determines acknowledgment strategy, Including:
If the emotional characteristics are negative emotional characteristics, and the degree value of the emotional characteristics is more than predetermined threshold value, and determination is answered It is dialogue heart-to-heart talk response mode to answer strategy;Or determine that acknowledgment strategy is first to be answered with dialogue heart-to-heart talk response mode It answers, then response is carried out with the audio resource broadcast mode.
10. according to the method described in claim 1, it is characterized in that, further including:
Determine user's portrait of the child user;Wherein, the child user portrait includes at least one of following characteristics, The attribute information of child user, the history intersection record of child user, child user habitual term, child user work and rest Incidence relation between rule, the audio resource of child user hobby, geographical location and child user;
It is drawn a portrait according to the user of the determining child user, the acknowledgment strategy is optimized.
11. according to the method described in claim 10, it is characterized in that, further including:
Obtain the temporal information and/or location information for receiving the interactive voice information;
According to the temporal information and/or location information, is drawn a portrait based on the user, determine that the child user is presently in field Scape;
It is presently in scene according to described, the acknowledgment strategy is optimized.
12. according to the method described in claim 1, it is characterized in that, further including:
According to predetermined period, the mood analysis report of the child user is generated.
13. a kind of interactive device based on children's mood, which is characterized in that including:
Receiving module, the interactive voice information for receiving child user;
Determining module, for determining interaction content in the interactive voice information;Youngster is determined in the interactive voice information The phonetic feature of virgin user;According to the interaction content, the emotional characteristics of the child user are determined;According to voice spy Sign, determines the degree value of the emotional characteristics;According to the emotional characteristics, the degree value of the emotional characteristics determines response plan Slightly.
14. device according to claim 13, which is characterized in that the determining module, including:
Transform subblock, for the interactive voice information to be converted to text message;
Submodule is analyzed, for carrying out semantic analysis to the text message;
Extracting sub-module obtains interaction content for extracting Topic word and/or emotion word in the text message;
Emotional characteristics determination sub-module, for being based on preset recognition rule, according to the Topic word and/or emotion word, Determine the emotional characteristics of the child user.
15. device according to claim 14, which is characterized in that further include:
Lexicon module establishes children's dictionary for being based on childrenese;
Wherein, the childrenese includes at least:The folded word representation language of children, animal are personalized representation language;The children Dictionary includes:The paraphrase of childrenese, the paraphrase of childrenese combination context, the mood mark of childrenese;
The mood mark of the childrenese is for identifying positive mood classification, negative emotions classification, neutral mood classification.
16. device according to claim 15, which is characterized in that
The emotional characteristics determination sub-module, be specifically used for determining in children's dictionary each Topic word and/or Each corresponding mood mark of the emotion word;
When not including negative emotions classification in each mood mark determined, and including positive mood classification, described in determination The emotional characteristics of child user are front emotional characteristics;
When not including positive mood classification in each mood mark determined, and including negative emotions classification, described in determination The emotional characteristics of child user are negative emotions feature;
When only including neutral mood classification in each mood mark determined, determine that the emotional characteristics of the child user are Neutral emotional characteristics;
When not only having included positive mood classification in each mood mark determined, but also when including negative emotions classification, according to described The word order of Topic word and/or emotion word in the interactive voice information is based on the context meaning of one's words, determines that the children use The emotional characteristics at family.
17. device according to claim 14, which is characterized in that further include:
Identification model module, for obtaining children's mood labeled data, training obtains children's Emotion identification model;
Correspondingly, the emotional characteristics determination sub-module, specifically for the Topic word and/or emotion word are input to institute In the children's Emotion identification model for stating identification model module, identification obtains the emotional characteristics of the child user.
18. device according to claim 14, which is characterized in that the determining module, including:
Phonetic feature determination sub-module, for determining at least one in following phonetic feature in the interactive voice information, Voice intensity, word speed, intonation.
19. device according to claim 14, which is characterized in that the determining module, including:
First degree value determination sub-module, for, integrally as objects of statistics, determining the voice using the interactive voice information The average value of the corresponding phonetic feature of interactive information;According to the average value of the phonetic feature, the emotional characteristics are determined Degree value.
20. device according to claim 14, which is characterized in that the determining module, including:
Second degree value determination sub-module, for determining in the interactive voice information each Topic word and/or described The phonetic feature of emotion word;According to the weighted value of different parts of speech, the phonetic feature of the interactive voice information is carried out Weighted calculation obtains the weighted average of the corresponding phonetic feature of the interactive voice information;According to the phonetic feature Weighted average, determine the degree value of the emotional characteristics.
21. device according to claim 13, which is characterized in that the acknowledgment strategy includes:Dialogue heart-to-heart talk response mode, And/or audio resource broadcast mode;The determining module, including:
First determination sub-module, for working as the emotional characteristics for negative emotional characteristics, and the degree value of the emotional characteristics When more than predetermined threshold value, determine that acknowledgment strategy is dialogue heart-to-heart talk response mode;Or determine that acknowledgment strategy is first with described Dialogue heart-to-heart talk response mode carries out response, then carries out response with the audio resource broadcast mode.
22. device according to claim 13, which is characterized in that the determining module further includes:
User's portrait determination sub-module, for determining that the user of the child user draws a portrait;Wherein, user portrait include with At least one of lower feature, the attribute information of child user, the history intersection record of child user, child user habituation Being associated between term, the work and rest rule of child user, the audio resource of child user hobby, geographical location and child user System;
Optimize submodule, the user of the child user for being determined according to user portrait determination sub-module draws a portrait, right The acknowledgment strategy optimizes.
23. device according to claim 22, which is characterized in that described device further includes:
Acquisition module, for obtaining the temporal information and/or location information for receiving the interactive voice information;
The determining module further includes:
Scene determination sub-module, for according to the temporal information and/or location information, being drawn a portrait based on the user, determining institute It states child user and is presently in scene;
The optimization submodule is additionally operable to be presently in scene according to, be optimized to the acknowledgment strategy.
24. device according to claim 13, which is characterized in that further include:
Generation module, for according to predetermined period, generating the mood analysis report of the child user.
25. a kind of electronic equipment, which is characterized in that including:Processor;Memory;And program;Wherein, described program is deposited Storage is configured as being executed by the processor in the memory, and described program includes for executing such as claim The instruction of 1-12 any one of them methods.
26. a kind of electronic equipment readable storage medium storing program for executing, which is characterized in that the electronic device-readable storage medium is stored with journey Sequence, described program make electronic equipment perform claim require 1-12 any one of them methods.
CN201810290987.3A 2018-03-30 2018-03-30 Interaction method and device based on child emotion Active CN108536802B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810290987.3A CN108536802B (en) 2018-03-30 2018-03-30 Interaction method and device based on child emotion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810290987.3A CN108536802B (en) 2018-03-30 2018-03-30 Interaction method and device based on child emotion

Publications (2)

Publication Number Publication Date
CN108536802A true CN108536802A (en) 2018-09-14
CN108536802B CN108536802B (en) 2020-01-14

Family

ID=63482290

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810290987.3A Active CN108536802B (en) 2018-03-30 2018-03-30 Interaction method and device based on child emotion

Country Status (1)

Country Link
CN (1) CN108536802B (en)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109243582A (en) * 2018-09-19 2019-01-18 江苏金惠甫山软件科技有限公司 The human-computer interaction motion management method and system of knowledge based graphical spectrum technology
CN109344229A (en) * 2018-09-18 2019-02-15 深圳壹账通智能科技有限公司 Method, apparatus, computer equipment and the storage medium of dialog analysis evaluation
CN109408708A (en) * 2018-09-25 2019-03-01 平安科技(深圳)有限公司 Method, apparatus, computer equipment and the storage medium that user recommends
CN109473122A (en) * 2018-11-12 2019-03-15 平安科技(深圳)有限公司 Mood analysis method, device and terminal device based on detection model
CN109547332A (en) * 2018-11-22 2019-03-29 腾讯科技(深圳)有限公司 Communication session interaction method and device, and computer equipment
CN109670166A (en) * 2018-09-26 2019-04-23 平安科技(深圳)有限公司 Collection householder method, device, equipment and storage medium based on speech recognition
CN109754810A (en) * 2019-02-21 2019-05-14 珠海格力电器股份有限公司 A kind of sound control method, device, storage medium and air-conditioning
CN109871807A (en) * 2019-02-21 2019-06-11 百度在线网络技术(北京)有限公司 Face image processing process and device
CN109885277A (en) * 2019-02-26 2019-06-14 百度在线网络技术(北京)有限公司 Human-computer interaction device, mthods, systems and devices
CN109935242A (en) * 2019-01-10 2019-06-25 上海言通网络科技有限公司 Formula speech processing system and method can be interrupted
CN109933782A (en) * 2018-12-03 2019-06-25 阿里巴巴集团控股有限公司 User emotion prediction technique and device
CN110246519A (en) * 2019-07-25 2019-09-17 深圳智慧林网络科技有限公司 Emotion identification method, equipment and computer readable storage medium
CN110264791A (en) * 2019-05-30 2019-09-20 合肥阿拉丁智能科技有限公司 Wrist-watch robot automtion autonomous operation system
CN110502609A (en) * 2019-07-11 2019-11-26 江苏心涧康科技有限公司 A kind of method, apparatus and company robot of adjusting mood
CN111048075A (en) * 2018-10-11 2020-04-21 上海智臻智能网络科技股份有限公司 Intelligent customer service system and intelligent customer service robot
CN111143529A (en) * 2019-12-24 2020-05-12 北京赤金智娱科技有限公司 Method and equipment for carrying out conversation with conversation robot
CN111696556A (en) * 2020-07-13 2020-09-22 上海茂声智能科技有限公司 Method, system, equipment and storage medium for analyzing user conversation emotion
WO2020253128A1 (en) * 2019-06-17 2020-12-24 深圳壹账通智能科技有限公司 Voice recognition-based communication service method, apparatus, computer device, and storage medium
CN112329431A (en) * 2019-08-01 2021-02-05 中国移动通信集团上海有限公司 Audio and video data processing method and device and storage medium
CN112017668B (en) * 2020-10-30 2021-09-24 北京淇瑀信息科技有限公司 Intelligent voice conversation method, device and system based on real-time emotion detection
CN113661036A (en) * 2019-04-16 2021-11-16 索尼集团公司 Information processing apparatus, information processing method, and program
CN113693600A (en) * 2021-08-27 2021-11-26 安徽淘云科技股份有限公司 Robot and method for psychology analysis of children based on behavior psychology
CN114566145A (en) * 2022-03-04 2022-05-31 河南云迹智能技术有限公司 Data interaction method, system and medium
WO2022126734A1 (en) * 2020-12-14 2022-06-23 美的集团股份有限公司 Voice interaction processing method and apparatus, electronic device, and storage medium
CN115460317A (en) * 2022-09-05 2022-12-09 西安万像电子科技有限公司 Emotion recognition and voice feedback method, device, medium and electronic equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110184721A1 (en) * 2006-03-03 2011-07-28 International Business Machines Corporation Communicating Across Voice and Text Channels with Emotion Preservation
CN103456314A (en) * 2013-09-03 2013-12-18 广州创维平面显示科技有限公司 Emotion recognition method and device
CN103593054A (en) * 2013-11-25 2014-02-19 北京光年无限科技有限公司 Question-answering system combining emotion recognition and output
CN104635574A (en) * 2014-12-15 2015-05-20 山东大学 Infant-oriented early-education accompanying and tending robot system
CN106683672A (en) * 2016-12-21 2017-05-17 竹间智能科技(上海)有限公司 Intelligent dialogue method and system based on emotion and semantics

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110184721A1 (en) * 2006-03-03 2011-07-28 International Business Machines Corporation Communicating Across Voice and Text Channels with Emotion Preservation
CN103456314A (en) * 2013-09-03 2013-12-18 广州创维平面显示科技有限公司 Emotion recognition method and device
CN103593054A (en) * 2013-11-25 2014-02-19 北京光年无限科技有限公司 Question-answering system combining emotion recognition and output
CN104635574A (en) * 2014-12-15 2015-05-20 山东大学 Infant-oriented early-education accompanying and tending robot system
CN106683672A (en) * 2016-12-21 2017-05-17 竹间智能科技(上海)有限公司 Intelligent dialogue method and system based on emotion and semantics

Cited By (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109344229A (en) * 2018-09-18 2019-02-15 深圳壹账通智能科技有限公司 Method, apparatus, computer equipment and the storage medium of dialog analysis evaluation
WO2020057014A1 (en) * 2018-09-18 2020-03-26 深圳壹账通智能科技有限公司 Dialogue analysis and evaluation method and apparatus, computer device and storage medium
CN109243582A (en) * 2018-09-19 2019-01-18 江苏金惠甫山软件科技有限公司 The human-computer interaction motion management method and system of knowledge based graphical spectrum technology
CN109408708A (en) * 2018-09-25 2019-03-01 平安科技(深圳)有限公司 Method, apparatus, computer equipment and the storage medium that user recommends
CN109670166A (en) * 2018-09-26 2019-04-23 平安科技(深圳)有限公司 Collection householder method, device, equipment and storage medium based on speech recognition
CN111048075A (en) * 2018-10-11 2020-04-21 上海智臻智能网络科技股份有限公司 Intelligent customer service system and intelligent customer service robot
CN109473122A (en) * 2018-11-12 2019-03-15 平安科技(深圳)有限公司 Mood analysis method, device and terminal device based on detection model
CN109547332A (en) * 2018-11-22 2019-03-29 腾讯科技(深圳)有限公司 Communication session interaction method and device, and computer equipment
CN109547332B (en) * 2018-11-22 2022-05-13 腾讯科技(深圳)有限公司 Communication session interaction method and device, and computer equipment
CN109933782B (en) * 2018-12-03 2023-11-28 创新先进技术有限公司 User emotion prediction method and device
CN109933782A (en) * 2018-12-03 2019-06-25 阿里巴巴集团控股有限公司 User emotion prediction technique and device
CN109935242A (en) * 2019-01-10 2019-06-25 上海言通网络科技有限公司 Formula speech processing system and method can be interrupted
CN109871807B (en) * 2019-02-21 2023-02-10 百度在线网络技术(北京)有限公司 Face image processing method and device
CN109871807A (en) * 2019-02-21 2019-06-11 百度在线网络技术(北京)有限公司 Face image processing process and device
CN109754810A (en) * 2019-02-21 2019-05-14 珠海格力电器股份有限公司 A kind of sound control method, device, storage medium and air-conditioning
CN109885277A (en) * 2019-02-26 2019-06-14 百度在线网络技术(北京)有限公司 Human-computer interaction device, mthods, systems and devices
CN113661036A (en) * 2019-04-16 2021-11-16 索尼集团公司 Information processing apparatus, information processing method, and program
CN110264791A (en) * 2019-05-30 2019-09-20 合肥阿拉丁智能科技有限公司 Wrist-watch robot automtion autonomous operation system
WO2020253128A1 (en) * 2019-06-17 2020-12-24 深圳壹账通智能科技有限公司 Voice recognition-based communication service method, apparatus, computer device, and storage medium
CN110502609A (en) * 2019-07-11 2019-11-26 江苏心涧康科技有限公司 A kind of method, apparatus and company robot of adjusting mood
CN110502609B (en) * 2019-07-11 2022-03-08 江苏心涧康科技有限公司 Method and device for adjusting emotion and accompanying robot
CN110246519A (en) * 2019-07-25 2019-09-17 深圳智慧林网络科技有限公司 Emotion identification method, equipment and computer readable storage medium
CN112329431A (en) * 2019-08-01 2021-02-05 中国移动通信集团上海有限公司 Audio and video data processing method and device and storage medium
CN112329431B (en) * 2019-08-01 2023-07-04 中国移动通信集团上海有限公司 Audio and video data processing method, equipment and storage medium
CN111143529A (en) * 2019-12-24 2020-05-12 北京赤金智娱科技有限公司 Method and equipment for carrying out conversation with conversation robot
CN111696556A (en) * 2020-07-13 2020-09-22 上海茂声智能科技有限公司 Method, system, equipment and storage medium for analyzing user conversation emotion
CN111696556B (en) * 2020-07-13 2023-05-16 上海茂声智能科技有限公司 Method, system, equipment and storage medium for analyzing user dialogue emotion
CN112017668B (en) * 2020-10-30 2021-09-24 北京淇瑀信息科技有限公司 Intelligent voice conversation method, device and system based on real-time emotion detection
WO2022126734A1 (en) * 2020-12-14 2022-06-23 美的集团股份有限公司 Voice interaction processing method and apparatus, electronic device, and storage medium
CN113693600A (en) * 2021-08-27 2021-11-26 安徽淘云科技股份有限公司 Robot and method for psychology analysis of children based on behavior psychology
CN114566145A (en) * 2022-03-04 2022-05-31 河南云迹智能技术有限公司 Data interaction method, system and medium
CN115460317A (en) * 2022-09-05 2022-12-09 西安万像电子科技有限公司 Emotion recognition and voice feedback method, device, medium and electronic equipment

Also Published As

Publication number Publication date
CN108536802B (en) 2020-01-14

Similar Documents

Publication Publication Date Title
CN108536802A (en) Exchange method based on children's mood and device
KR102627948B1 (en) Automated assistants that accommodate multiple age groups and/or vocabulary levels
US11645547B2 (en) Human-machine interactive method and device based on artificial intelligence
US6795808B1 (en) User interface/entertainment device that simulates personal interaction and charges external database with relevant data
Kurumada et al. Pragmatic interpretation of contrastive prosody: It looks like speech adaptation
Berg Modelling of natural dialogues in the context of speech-based information and control systems
CN106653052A (en) Virtual human face animation generation method and device
CN109189980A (en) The method and electronic equipment of interactive voice are carried out with user
KR20020067592A (en) User interface/entertainment device that simulates personal interaction and responds to user's mental state and/or personality
KR20020067590A (en) Environment-responsive user interface/entertainment device that simulates personal interaction
KR20020067591A (en) Self-updating user interface/entertainment device that simulates personal interaction
KR20210070213A (en) Voice user interface
CN111145721A (en) Personalized prompt language generation method, device and equipment
KR20040014123A (en) Emotion and Motion Extracting System of Virtual Human and its Method
Pittermann et al. Handling emotions in human-computer dialogues
CN108986785A (en) A kind of text adaptation method and device
KR102101311B1 (en) Method and apparatus for providing virtual reality including virtual pet
Mairesse Learning to adapt in dialogue systems: data-driven models for personality recognition and generation.
Edlund In search for the conversational homunculus: serving to understand spoken human face-to-face interaction
Padilha Modelling turn-taking in a simulation of small group discussion
JP6930781B1 (en) Learning method and content playback device
JP2019139282A (en) Interaction control device capable of carrying out interaction about contents, program and method
Missaouib et al. Interactive Storytelling for Children: A Case-study of Design and Development Considerations for Ethical Conversational AI
Simeoni et al. EMPATHIC VOICE: ENABLING EMOTIONAL INTELLIGENCE IN VIRTUAL ASSISTANTS

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20210518

Address after: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing

Patentee after: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd.

Patentee after: Shanghai Xiaodu Technology Co.,Ltd.

Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing

Patentee before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd.