CN112565663A - Demand question reply method and device, terminal equipment and storage medium - Google Patents

Demand question reply method and device, terminal equipment and storage medium Download PDF

Info

Publication number
CN112565663A
CN112565663A CN202011348081.6A CN202011348081A CN112565663A CN 112565663 A CN112565663 A CN 112565663A CN 202011348081 A CN202011348081 A CN 202011348081A CN 112565663 A CN112565663 A CN 112565663A
Authority
CN
China
Prior art keywords
voice
candidate
clarified
demand
user
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
CN202011348081.6A
Other languages
Chinese (zh)
Other versions
CN112565663B (en
Inventor
张婷
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.)
Ping An Puhui Enterprise Management Co Ltd
Original Assignee
Ping An Puhui Enterprise Management 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 Ping An Puhui Enterprise Management Co Ltd filed Critical Ping An Puhui Enterprise Management Co Ltd
Priority to CN202011348081.6A priority Critical patent/CN112565663B/en
Publication of CN112565663A publication Critical patent/CN112565663A/en
Application granted granted Critical
Publication of CN112565663B publication Critical patent/CN112565663B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/14—Systems for two-way working
    • H04N7/15—Conference systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/02—Methods for producing synthetic speech; Speech synthesisers
    • G10L13/04—Details of speech synthesis systems, e.g. synthesiser structure or memory management
    • G10L13/047—Architecture of speech synthesisers
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/18—Speech classification or search using natural language modelling
    • G10L15/1822—Parsing for meaning understanding
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • G10L25/54—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for retrieval

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Signal Processing (AREA)
  • Artificial Intelligence (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application is suitable for the technical field of artificial intelligence, and provides a demand question replying method, a demand question replying device, terminal equipment and a storage medium, wherein the method comprises the following steps: performing semantic analysis on the demand problem to obtain semantic keywords; performing voice query according to the semantic keywords to obtain candidate clear voice; screening the candidate clear voice according to the user information to obtain voice information of the screened candidate clear voice; calculating an association score between the candidate clarified voice and the user according to the voice recording time and the voice tag information of the candidate clarified voice; and replying the question to the demand question by the candidate clarified voice corresponding to the maximum association score. According to the method and the device, the association degree between different candidate clarified voices and the user can be effectively calculated according to the voice recording time and the voice tag information of the candidate clarified voices, and the problem is answered to the demand problem through the candidate clarified voices corresponding to the maximum association score, so that the demand problem is automatically answered. In addition, the application also relates to a block chain technology.

Description

Demand question reply method and device, terminal equipment and storage medium
Technical Field
The present application relates to the field of artificial intelligence, and in particular, to a demand problem replying method, apparatus, terminal device, and storage medium.
Background
At present, the demand problems are communicated and clarified in a video conference mode, when some people do not participate in the video conference, the same demand problems need to be repeatedly replied, and then the communication cost of the demand problems is increased.
In the existing demand problem reply process, oral transfer is carried out in a manual mode, so that personnel in the field can not obtain the clarified content of the corresponding demand problem, but deviation exists in oral reply in a manual mode, and the accuracy of demand problem reply is further reduced.
Disclosure of Invention
In view of this, embodiments of the present application provide a method and an apparatus for replying a demand problem, a terminal device, and a storage medium, so as to solve the problem of low accuracy of replying a demand problem in the prior art, which is caused by manually speaking the demand problem in a demand problem replying process.
A first aspect of an embodiment of the present application provides a demand problem replying method, including:
acquiring a demand problem sent by a user, and performing semantic analysis on the demand problem to obtain semantic keywords;
performing voice query according to the semantic keywords to obtain candidate clear voice and acquire user information of the user;
screening the candidate clarified voice according to the user information, and acquiring voice information of the screened candidate clarified voice, wherein the voice information comprises voice recording time and voice label information;
calculating an association score between the candidate clarified voice and the user according to the voice recording time and the voice tag information of the candidate clarified voice after screening, wherein the association score is used for representing the degree of association between the corresponding candidate clarified voice and the user;
and replying the question to the demand question by the candidate clarified voice corresponding to the maximum association score.
Further, after calculating the association score between the candidate clarified speech and the user according to the speech recording time and the speech tag information of the candidate clarified speech after the filtering, the method further includes:
if the maximum correlation score is smaller than a score threshold value, inquiring a target clarification object according to the semantic keywords, wherein the target clarification object is a user preset aiming at the semantic keywords;
establishing a clarification conference aiming at the demand problem, and sending a conference request to the inquired target clarification object;
and after the target clarification object joins the clarification conference, acquiring reply voice of the target clarification object to the demand problem in the clarification conference, and performing problem reply on the demand problem by using the reply voice.
Further, the establishing a clarification conference for the demand problem and sending a conference request to the inquired target clarification object includes:
establishing conference group chat aiming at the demand question, and carrying out name marking on the conference group chat according to the semantic keyword corresponding to the demand question;
acquiring a group chat address of the conference group chat, and generating the conference request according to the group chat address and the name label of the conference group chat;
and sending the meeting request to the inquired target clarification object.
Further, the screening the candidate clarified speech according to the user information includes:
acquiring a conference participation record in the user information, wherein the conference participation record stores an identifier of a clarification conference participated by the user;
respectively obtaining conference identifications of different candidate clarified voices, and performing identification matching on the conference identifications of the candidate clarified voices and the conference participation records;
if the conference identification of the candidate clarified voice does not match with the conference participation record, retaining the candidate clarified voice;
and if the conference identification of the candidate clarified voice is matched with the conference participation record, deleting the candidate clarified voice.
Further, the semantic analysis of the requirement problem to obtain semantic keywords includes:
performing word segmentation on the demand problem to obtain word segmentation vocabularies, and combining different word segmentation vocabularies to obtain combined vocabularies;
respectively obtaining vocabulary association degrees corresponding to the combined vocabulary, wherein the vocabulary association degrees are used for representing association degrees between different word segmentation vocabularies in the combined vocabulary;
and setting the combined vocabulary corresponding to the maximum vocabulary association degree as the semantic keywords of the requirement problem.
Further, the calculating an association score between the candidate clarified speech and the user according to the speech recording time and the speech tag information of the candidate clarified speech after the screening includes:
acquiring a user tag in the user information, and performing tag matching on the user tag and the voice tag information to obtain a tag matching rate;
acquiring query time of the demand problem, and determining a scoring coefficient according to the query time and the voice recording time;
and performing score calculation on the label matching rate according to the score coefficient to obtain the association score.
Further, the obtaining of the reply voice of the target clarification object to the demand question at the clarification meeting comprises:
if any target clarifying object replies to the demand problem in the clarifying meeting, acquiring reply information of the target clarifying object to the demand problem in the clarifying meeting;
and carrying out voice conversion on the acquired reply information to obtain the reply voice.
A second aspect of an embodiment of the present application provides a demand issue reply device, including:
the semantic analysis unit is used for acquiring a demand problem sent by a user and performing semantic analysis on the demand problem to obtain semantic keywords;
the voice query unit is used for performing voice query according to the semantic keywords to obtain candidate clear voice and acquiring user information of the user;
the voice screening unit is used for screening the candidate clarified voice according to the user information and acquiring the voice information of the screened candidate clarified voice, wherein the voice information comprises voice recording time and voice label information;
the association score calculation unit is used for calculating an association score between the candidate clarified voice and the user according to the voice recording time and the voice tag information of the filtered candidate clarified voice, and the association score is used for representing the degree of association between the corresponding candidate clarified voice and the user;
and the question replying unit is used for replying the question to the demand question by the candidate clarified voice corresponding to the maximum association score.
A third aspect of the embodiments of the present application provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the terminal device, where the processor implements the steps of the demand problem recovery method provided by the first aspect when executing the computer program.
A fourth aspect of the embodiments of the present application provides a storage medium, where a computer program is stored, and the computer program, when executed by a processor, implements the steps of the demand problem recovery method provided by the first aspect.
The demand problem replying method, the demand problem replying device, the terminal equipment and the storage medium have the following advantages that: the method comprises the steps of obtaining semantic keywords capable of representing the speech of the demand problem by performing semantic analysis on the demand problem so as to improve the accuracy of query of candidate clarified speech, obtaining user information of a user, screening the candidate clarified speech according to the user information so as to delete the speech which is not matched with the user in the candidate clarified speech, improving the accuracy of subsequent question reply on the demand problem, calculating association scores between the candidate clarified speech and the user according to the speech tag information and the speech recording time of the screened candidate clarified speech so as to calculate the association degrees between different candidate clarified speech and the user, and performing question reply on the demand problem by the candidate clarified speech corresponding to the maximum association score so as to achieve the effect of automatically replying the demand problem and prevent deviation caused by manual dictation, the accuracy of demand problem recovery is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a flowchart illustrating an implementation of a method for replying to a demand issue according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating an implementation of a method for replying to a demand issue according to another embodiment of the present application;
FIG. 3 is a block diagram illustrating a structure of a demand question replying device according to an embodiment of the present disclosure;
fig. 4 is a block diagram of a terminal device according to an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The demand problem replying method according to the embodiment of the present application may be executed by a control device or a terminal (hereinafter referred to as a "mobile terminal").
Referring to fig. 1, fig. 1 is a flowchart illustrating an implementation of a demand question replying method according to an embodiment of the present application, including:
and step S10, acquiring a demand question sent by a user, and performing semantic analysis on the demand question to obtain a semantic keyword.
Optionally, in this step, performing semantic analysis on the demand problem to obtain a semantic keyword representing semantics of the demand problem, and obtaining the semantic keyword includes:
performing word segmentation on the demand problem to obtain word segmentation vocabularies, and combining different word segmentation vocabularies to obtain combined vocabularies;
respectively obtaining vocabulary association degrees corresponding to the combined vocabularies, and setting the combined vocabularies corresponding to the maximum vocabulary association degrees as the semantic keywords of the demand problem;
the requirement problem can be segmented based on a preset vocabulary, preset appointed vocabularies are stored in the preset vocabulary, the requirement problem is matched with the appointed vocabularies in the preset vocabulary, the requirement problem is segmented according to a matching result between the appointed vocabularies, the requirement problem is segmented, the segmentation vocabularies are obtained by segmenting the requirement problem, and vocabulary combination between the segmentation vocabularies is effectively facilitated.
Specifically, in this step, the vocabulary association degree is used to represent the association degree between different segmented vocabularies in the combined vocabulary, and the combined vocabulary is obtained by combining the different segmented vocabularies, for example, the segmented vocabulary a1, the segmented vocabulary a2, the segmented vocabulary a3 and the segmented vocabulary a4 are obtained after the requirement question is segmented, so that the combined vocabulary obtained by combining includes the combined vocabulary a1 a2, the combined vocabulary a1 a3, the combined vocabulary a1 a4, the combined vocabulary a2 a3, the combined vocabulary a2 a4, the combined vocabulary a3 a4, the combined vocabulary a1 a2 a3, the combined vocabulary a1 a2 a4 and the combined vocabulary a2 a3 a 4.
In addition, in the present embodiment, an association degree lookup table is pre-stored, in which the corresponding relationship between different combination vocabularies and the corresponding vocabulary association degrees is stored, therefore, the combined vocabulary obtained by combination is respectively matched with the association degree lookup table to obtain the corresponding vocabulary association degree, in this step, the vocabulary association degrees corresponding to the combined vocabulary are respectively obtained to obtain semantic keywords representing the semantics of the requirement problem, for example, when the combination vocabulary corresponding to the requirement problem is the combination vocabulary a1 a2, the combination vocabulary a1 a3 and the combination vocabulary a1 a4, and the vocabulary association degree corresponding to the combined vocabulary a1 a2 is greater than the vocabulary association degree corresponding to the combined vocabulary a1 a3, and the vocabulary association degree corresponding to the combined vocabulary a1 a3 is greater than the vocabulary association degree corresponding to the combined vocabulary a2 a3, then the combined vocabulary a1 a2 is set as the language keyword of the requirement problem.
And step S20, performing voice query according to the semantic keywords to obtain candidate clear voice, and acquiring the user information of the user.
In this embodiment, a speech clarification database is prestored, and a corresponding relationship between different semantic keywords and corresponding candidate clarified speech is stored in the speech clarification database, so that the speech keyword corresponding to the demand problem is matched with the speech clarification database to obtain corresponding candidate clarified speech, which is a reply speech of a clarified object on a corresponding clarified conference to the corresponding demand problem.
Optionally, the semantic keywords in the speech clarification database and the clarified speech may be stored in a one-to-one or many-to-one manner, and after performing semantic analysis on a requirement problem in step S1, a plurality of different semantic keywords may be obtained, for example, when the combined vocabulary corresponding to the requirement problem is the combined vocabulary a1 a2, the combined vocabulary a1 a3, and the combined vocabulary a1 a4, and the vocabulary association degrees corresponding to the combined vocabulary a1 a2 and the combined vocabulary a1 a3 are both greater than the association degree threshold, the combined vocabulary a1 a2 and the combined vocabulary a1 a3 are both set as the semantic keywords of the requirement problem.
And step S30, screening the candidate clarified voice according to the user information, and acquiring the voice information of the screened candidate clarified voice.
The user information comprises a conference participation record of the user, the identification and the conference time of the clarification conference participated by the user are stored in the conference participation record, and the accuracy of obtaining the voice information of the candidate clarification voice is effectively improved by screening the candidate clarification voice according to the user information.
For example, when it is determined that the user has a clear conference corresponding to the candidate clear voice according to the conference participation record, the voice information of the candidate clear voice does not need to be acquired, so that the accuracy of acquiring the voice information of the candidate clear voice is improved.
Specifically, in this step, the voice information includes a voice recording time and voice tag information, and the voice tag information includes at least one voice tag, and the voice tag is used for characterizing the type of the content of the voice information, for example, the voice tag may be "sports", "education", or "movie".
Step S40, calculating the association score between the candidate clarified voice and the user according to the voice recording time and the voice label information of the candidate clarified voice after screening.
The association score is used for representing the association degree between the corresponding candidate clarified voice and the user, when the association score between the candidate clarified voice and the user is larger, the association degree between the candidate clarified voice and the requirement problem sent by the user is larger, and when the association score between the candidate clarified voice and the user is smaller, the association degree between the candidate clarified voice and the requirement problem sent by the user is smaller.
Optionally, in this step, the user tag of the user is determined according to the user information of the user, if the user tag is more similar to the voice tag information, the association score between the voice tag information and the user is larger, and if the user tag is less similar to the voice tag information, the association score between the voice tag information and the user is smaller.
Further, in this step, the user tag of the user may be determined by obtaining information of the gender, the marital status, the age, the home address, the occupation, and the like of the user, for example, when the gender of the user is male and the marital status is married, it may be determined that the user carries a "father" tag, and when the occupation of the user is a math teacher, it may be determined that the user carries a "math" tag.
Further, in this embodiment, the similarity between the user tag and the voice tag information may be calculated by using a euclidean distance formula.
Specifically, in this step, the calculating a correlation score between the candidate clarified speech and the user according to the speech recording time and the speech tag information of the candidate clarified speech after the screening includes:
acquiring a user tag in the user information, and performing tag matching on the user tag and the voice tag information to obtain a tag matching rate;
acquiring query time of the demand problem, and determining a scoring coefficient according to the query time and the voice recording time;
performing score calculation on the tag matching rate according to the score coefficient to obtain the association score;
the method comprises the steps of obtaining a user label in user information, carrying out label matching on the user label and voice label information, effectively calculating the label matching rate between a user and corresponding candidate clarified voice, and carrying out coefficient weighting calculation on the label matching rate by using a score coefficient to obtain a corresponding association score.
Specifically, in this step, the score coefficient is obtained by respectively calculating a time difference between the query time and the voice recording time, and matching the time difference with a pre-stored coefficient lookup table, where the coefficient lookup table stores corresponding relationships between different time differences and corresponding score coefficients.
Optionally, in this step, coefficient weighting calculation may be performed between the score coefficient and the tag matching rate by using a multiplication method, for example, when the determined score coefficient is 1.5 and the tag matching rate between the user tag and the voice tag information is 0.8, the corresponding association score is 0.8 × 1.5 — 1.2.
And step S50, performing question reply on the demand question by the candidate clarified voice corresponding to the maximum association score.
Wherein, carry out answer display through to this demand question to reach the effect of answering the question to the demand question.
Optionally, in this step, if a response error instruction of the user for the candidate clarified speech corresponding to the maximum association score is received, the association scores with the maximum association score removed are sorted to obtain a score sorting table, and the candidate clarified speech in the preset sorting in the score sorting table is subjected to question response on the demand question.
Specifically, in this step, when the association scores without the maximum association score are sorted in the forward order, the question of the demand question is answered by the candidate clarified voices in the preset order in the front of the score sorting table, and when the association scores without the maximum association score are sorted in the reverse order, the question of the demand question is answered by the candidate clarified voices in the preset order in the back of the score sorting table, optionally, the preset order may be set according to the demand, for example, the preset order may be set to 3, 5, or 10, and the like.
In the embodiment, semantic analysis is performed on the demand problem to obtain semantic keywords capable of representing the voice of the demand problem, so that the accuracy of query of the candidate clarified voice is improved, the user information of the user is obtained, the candidate clarified voice is screened according to the user information to delete the voice which is not matched with the user in the candidate clarified voice, so that the accuracy of subsequent problem response to the demand problem is improved, the association score between the candidate clarified voice and the user is calculated according to the voice tag information and the voice recording time of the screened candidate clarified voice, so as to calculate the association degree between different candidate clarified voices and the user, the problem response is performed on the demand problem by the candidate clarified voice corresponding to the maximum association score, so that the effect of automatically responding to the demand problem is achieved, and the deviation caused by manual dictation is prevented, the accuracy of demand problem recovery is improved.
Referring to fig. 2, fig. 2 is a flowchart illustrating an implementation of a method for replying to a demand issue according to another embodiment of the present application. With respect to the embodiment corresponding to fig. 1, after step S40, the method for replying to a demand issue further includes:
and step S60, if the maximum associated score is smaller than a score threshold, querying a target clarification object according to the semantic keywords.
The score threshold may be set according to a requirement, for example, the score threshold may be set to 0.5, 1, or 1.5, and the score threshold is used to determine whether the candidate clarified speech corresponding to the maximum association score satisfies the question answering condition.
Specifically, in this step, when the maximum association score is smaller than the score threshold, it is determined that the candidate clarified speech corresponding to the maximum association score does not satisfy the question reply condition, and when the maximum association score is greater than or equal to the score threshold, it is determined that the candidate clarified speech corresponding to the maximum association score satisfies the question reply condition, and the question reply may be performed on the demand question according to the candidate clarified speech corresponding to the maximum association score.
In this step, the target clarifying object is a user preset for the semantic keyword, an object lookup table is prestored in this embodiment, and a corresponding relationship between different semantic keywords and the corresponding target clarifying object is stored in the object lookup table, so that the corresponding target clarifying object is obtained by matching the semantic keyword with the object lookup table.
And step S70, establishing a clarification conference aiming at the requirement problem, and sending a conference request to the inquired target clarification object.
In this embodiment, a clarification conference establishing program is prestored, and when it is determined in step S60 that the maximum associated score is smaller than the score threshold, the clarification conference establishing program is triggered to establish a clarification conference for the demand issue, where the clarification conference is used to provide an online reply function for the demand issue.
In this step, a conference request is sent to the inquired target clarifying object to prompt the target clarifying object to join in the corresponding clarifying conference, the conference request stores a conference address corresponding to the clarifying conference, the conference address can be a connection address corresponding to any group chat, and the conference address is used for guaranteeing the connection between the target clarifying object and the clarifying conference.
Specifically, in this step, the establishing a clarification conference for the demand problem and sending a conference request to the inquired target clarification object includes:
establishing conference group chat aiming at the demand question, and carrying out name marking on the conference group chat according to the semantic keyword corresponding to the demand question;
acquiring a group chat address of the conference group chat, and generating the conference request according to the group chat address and the name label of the conference group chat;
sending the meeting request to the inquired target clarification object;
the name of the conference group chat can be tagged in a text or digital manner, for example, when the semantic keyword corresponding to the requirement question is "sports", the name of the conference group chat is tagged according to the "sports".
Specifically, in the step, the conference request is generated by marking the acquired group chat address and the name of the conference group chat, and is sent to the inquired target clarification object, so that the connection between the target clarification object and the clarification conference is effectively guaranteed, and the accuracy of the target clarification object in responding to the demand problem is improved.
Step S80, after the target clarification object joins the clarification conference, acquiring the reply voice of the target clarification object to the requirement question in the clarification conference, and replying the reply voice to the requirement question.
Wherein, through obtaining the reply pronunciation of target clarification object to the demand problem in the clarification meeting to reply pronunciation and carry out the problem reply to the demand problem, can adopt the mode that the online problem replied to directly reply this demand problem, need not to carry out the answer of demand problem based on the mode of artifical dictation, improved the accuracy that the demand problem replied.
Optionally, in this step, the obtaining of the reply voice of the target clarification object to the demand problem in the clarification conference includes:
if any target clarifying object replies to the demand problem in the clarifying meeting, acquiring reply information of the target clarifying object to the demand problem in the clarifying meeting, and performing voice conversion on the acquired reply information to obtain the reply voice;
the voice conversion is carried out on the acquired reply information, so that the text information can be effectively and automatically converted into the voice information, the answering of the answer corresponding to the demand problem by the user is effectively facilitated, and the use experience of the user is improved.
Further, in this embodiment, with respect to step S30 in the embodiment of fig. 1, the filtering the candidate clarified speech according to the user information includes:
acquiring a conference participation record in the user information, wherein the conference participation record stores an identifier of a clarification conference participated by the user;
respectively obtaining conference identifications of different candidate clarified voices, and performing identification matching on the conference identifications of the candidate clarified voices and the conference participation records, wherein whether a user participates in a clarified conference corresponding to the conference identification is judged by performing identification matching on the conference identification of the selected clarified voice and the conference participation records;
if the conference identification of the candidate clarified voice is not matched with the conference participation record, the candidate clarified voice is reserved, and if the conference identification of the candidate clarified voice is not matched with the conference participation record, the user is judged not to participate in the clarified conference corresponding to the conference identification;
and if the conference identifier of the candidate clarified voice is matched with the conference participation record, deleting the candidate clarified voice, and if the conference identifier of the candidate clarified voice is matched with the conference participation record, judging that the user already participates in the clarified conference corresponding to the conference identifier, so that the candidate clarified voice can be effectively screened by deleting the candidate clarified voice corresponding to the conference identifier.
In the embodiment, the target clarifying object is inquired according to the semantic keyword, the conference request is sent to the inquired target clarifying object, the target clarifying object can be effectively prompted to be added into the clarifying conference, the problem reply is carried out on the demand problem corresponding to the clarifying conference, the reply voice of the target clarifying object to the demand problem on the clarifying conference is obtained, the problem reply is carried out on the demand problem by the reply voice, the problem reply to the demand problem under the condition that the candidate clarifying voice corresponding to the maximum association score does not meet the problem reply condition is guaranteed, and the accuracy of the demand problem reply is improved.
In all embodiments of the present application, the association score between the candidate clarified speech and the user is obtained based on the speech recording time and the speech tag information of the candidate clarified speech after the screening, and specifically, the association score between the candidate clarified speech and the user is obtained from the speech recording time and the speech tag information of the candidate clarified speech after the screening. Uploading the association score between the candidate clarified speech and the user to the blockchain may ensure its security and fair transparency to the user. The user device may download the association score between the candidate clarified speech and the user from the blockchain to verify whether the association score between the candidate clarified speech and the user is tampered. The blockchain referred to in this example is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Referring to fig. 3, fig. 3 is a block diagram illustrating a demand problem recovery apparatus 100 according to an embodiment of the present disclosure. The demand question recovery apparatus 100 in this embodiment includes units for executing the steps in the embodiments corresponding to fig. 1 and fig. 2. Please refer to fig. 1 and fig. 2 and the related descriptions in the embodiments corresponding to fig. 1 and fig. 2. For convenience of explanation, only the portions related to the present embodiment are shown. Referring to fig. 3, the demand question restoration apparatus 100 includes: a semantic analysis unit 10, a voice query unit 11, a voice screening unit 12, an association score calculation unit 13, and a question reply unit 14, wherein:
and the semantic analysis unit 10 is configured to acquire a demand question sent by a user, and perform semantic analysis on the demand question to obtain a semantic keyword.
Wherein, the semantic analysis unit 10 is further configured to: performing word segmentation on the demand problem to obtain word segmentation vocabularies, and combining different word segmentation vocabularies to obtain combined vocabularies;
respectively obtaining vocabulary association degrees corresponding to the combined vocabulary, wherein the vocabulary association degrees are used for representing association degrees between different word segmentation vocabularies in the combined vocabulary;
and setting the combined vocabulary corresponding to the maximum vocabulary association degree as the semantic keywords of the requirement problem.
And the voice query unit 11 is configured to perform voice query according to the semantic keywords to obtain candidate clarified voices, and obtain user information of the user.
And the voice screening unit 12 is configured to screen the candidate clarified voice according to the user information, and acquire voice information of the screened candidate clarified voice, where the voice information includes voice recording time and voice tag information.
And the association score calculation unit 13 is configured to calculate an association score between the candidate clarified speech and the user according to the speech recording time and the speech tag information of the filtered candidate clarified speech, where the association score is used to characterize an association degree between the corresponding candidate clarified speech and the user.
Wherein, the association score calculating unit 13 is further configured to: acquiring a user tag in the user information, and performing tag matching on the user tag and the voice tag information to obtain a tag matching rate;
acquiring query time of the demand problem, and determining a scoring coefficient according to the query time and the voice recording time;
and performing score calculation on the label matching rate according to the score coefficient to obtain the association score.
And a question replying unit 14, configured to perform question replying on the demand question by the candidate clarified speech corresponding to the largest association score.
Optionally, the demand question recovery apparatus 100 further includes:
a clarifying meeting establishing unit 15, configured to query a target clarifying object according to the semantic keyword if the maximum associated score is smaller than a score threshold, where the target clarifying object is a user preset for the semantic keyword;
establishing a clarification conference aiming at the demand problem, and sending a conference request to the inquired target clarification object;
and after the target clarification object joins the clarification conference, acquiring reply voice of the target clarification object to the demand problem in the clarification conference, and performing problem reply on the demand problem by using the reply voice.
Wherein the clarifying meeting establishing unit 15 is further configured to: establishing conference group chat aiming at the demand question, and carrying out name marking on the conference group chat according to the semantic keyword corresponding to the demand question;
acquiring a group chat address of the conference group chat, and generating the conference request according to the group chat address and the name label of the conference group chat;
and sending the meeting request to the inquired target clarification object.
Further, the clarifying meeting establishing unit 15 is further configured to: if any target clarifying object replies to the demand problem in the clarifying meeting, acquiring reply information of the target clarifying object to the demand problem in the clarifying meeting;
and carrying out voice conversion on the acquired reply information to obtain the reply voice.
Optionally, in this embodiment, the voice screening unit 12 is further configured to: acquiring a conference participation record in the user information, wherein the conference participation record stores an identifier of a clarification conference participated by the user;
respectively obtaining conference identifications of different candidate clarified voices, and performing identification matching on the conference identifications of the candidate clarified voices and the conference participation records;
if the conference identification of the candidate clarified voice does not match with the conference participation record, retaining the candidate clarified voice;
and if the conference identification of the candidate clarified voice is matched with the conference participation record, deleting the candidate clarified voice.
In the embodiment, semantic analysis is performed on the demand problem to obtain semantic keywords capable of representing the voice of the demand problem, so that the accuracy of query of the candidate clarified voice is improved, the user information of the user is obtained, the candidate clarified voice is screened according to the user information to delete the voice which is not matched with the user in the candidate clarified voice, so that the accuracy of subsequent problem response to the demand problem is improved, the association score between the candidate clarified voice and the user is calculated according to the voice tag information and the voice recording time of the screened candidate clarified voice, so as to calculate the association degree between different candidate clarified voices and the user, the problem response is performed on the demand problem by the candidate clarified voice corresponding to the maximum association score, so that the effect of automatically responding to the demand problem is achieved, and the deviation caused by manual dictation is prevented, the accuracy of demand problem recovery is improved.
Fig. 4 is a block diagram of a terminal device 2 according to another embodiment of the present application. As shown in fig. 4, the terminal device 2 of this embodiment includes: a processor 20, a memory 21 and a computer program 22 stored in said memory 21 and executable on said processor 20, such as a program of a demand question answering method. The processor 20, when executing the computer program 23, implements the steps of the above-mentioned requirement problem recovery method in each embodiment, such as S10-S50 shown in fig. 1 or S60-S80 shown in fig. 2. Alternatively, when the processor 20 executes the computer program 22, the functions of the units in the embodiment corresponding to fig. 3, for example, the functions of the units 10 to 15 shown in fig. 3, are implemented, for which reference is specifically made to the relevant description in the embodiment corresponding to fig. 4, which is not repeated herein.
Illustratively, the computer program 22 may be divided into one or more units, which are stored in the memory 21 and executed by the processor 20 to accomplish the present application. The one or more units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution of the computer program 22 in the terminal device 2. For example, the computer program 22 may be divided into a semantic analysis unit 10, a voice query unit 11, a voice screening unit 12, an association score calculation unit 13, a question answering unit 14, and a clarification conference establishing unit 15, each of which functions as described above.
The terminal device may include, but is not limited to, a processor 20, a memory 21. It will be appreciated by those skilled in the art that fig. 4 is merely an example of a terminal device 2 and does not constitute a limitation of the terminal device 2 and may include more or less components than those shown, or some components may be combined, or different components, for example the terminal device may also include input output devices, network access devices, buses, etc.
The Processor 20 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 21 may be an internal storage unit of the terminal device 2, such as a hard disk or a memory of the terminal device 2. The memory 21 may also be an external storage device of the terminal device 2, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the terminal device 2. Further, the memory 21 may also include both an internal storage unit and an external storage device of the terminal device 2. The memory 21 is used for storing the computer program and other programs and data required by the terminal device. The memory 21 may also be used to temporarily store data that has been output or is to be output.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims (10)

1. A method for demand problem recovery, comprising:
acquiring a demand problem sent by a user, and performing semantic analysis on the demand problem to obtain semantic keywords;
performing voice query according to the semantic keywords to obtain candidate clear voice and acquire user information of the user;
screening the candidate clarified voice according to the user information, and acquiring voice information of the screened candidate clarified voice, wherein the voice information comprises voice recording time and voice label information;
calculating an association score between the candidate clarified voice and the user according to the voice recording time and the voice tag information of the candidate clarified voice after screening, wherein the association score is used for representing the degree of association between the corresponding candidate clarified voice and the user;
and replying the question to the demand question by the candidate clarified voice corresponding to the maximum association score.
2. The demand question answering method according to claim 1, wherein after calculating an association score between the candidate clarified speech and the user based on the speech recording time and the speech tag information of the filtered candidate clarified speech, the method further comprises:
if the maximum correlation score is smaller than a score threshold value, inquiring a target clarification object according to the semantic keywords, wherein the target clarification object is a user preset aiming at the semantic keywords;
establishing a clarification conference aiming at the demand problem, and sending a conference request to the inquired target clarification object;
and after the target clarification object joins the clarification conference, acquiring reply voice of the target clarification object to the demand problem in the clarification conference, and performing problem reply on the demand problem by using the reply voice.
3. The demand question replying method according to claim 2, wherein the establishing of the clarification meeting for the demand question and the sending of the meeting request to the inquired target clarification object comprises:
establishing conference group chat aiming at the demand question, and carrying out name marking on the conference group chat according to the semantic keyword corresponding to the demand question;
acquiring a group chat address of the conference group chat, and generating the conference request according to the group chat address and the name label of the conference group chat;
and sending the meeting request to the inquired target clarification object.
4. The method for answering the demand question according to claim 2, wherein the screening the candidate clarified voices according to the user information includes:
acquiring a conference participation record in the user information, wherein the conference participation record stores an identifier of a clarification conference participated by the user;
respectively obtaining conference identifications of different candidate clarified voices, and performing identification matching on the conference identifications of the candidate clarified voices and the conference participation records;
if the conference identification of the candidate clarified voice does not match with the conference participation record, retaining the candidate clarified voice;
and if the conference identification of the candidate clarified voice is matched with the conference participation record, deleting the candidate clarified voice.
5. The demand question replying method according to claim 1, wherein the semantic analyzing the demand question to obtain semantic keywords comprises:
performing word segmentation on the demand problem to obtain word segmentation vocabularies, and combining different word segmentation vocabularies to obtain combined vocabularies;
respectively obtaining vocabulary association degrees corresponding to the combined vocabulary, wherein the vocabulary association degrees are used for representing association degrees between different word segmentation vocabularies in the combined vocabulary;
and setting the combined vocabulary corresponding to the maximum vocabulary association degree as the semantic keywords of the requirement problem.
6. The demand question answering method according to claim 1, wherein the calculating of the association score between the candidate clarified speech and the user based on the speech recording time and the speech tag information of the candidate clarified speech after the filtering includes:
acquiring a user tag in the user information, and performing tag matching on the user tag and the voice tag information to obtain a tag matching rate;
acquiring query time of the demand problem, and determining a scoring coefficient according to the query time and the voice recording time;
and performing score calculation on the label matching rate according to the score coefficient to obtain the association score.
7. The demand question replying method according to claim 2, wherein the obtaining of the reply voice of the target clarification object to the demand question at the clarification conference comprises:
if any target clarifying object replies to the demand problem in the clarifying meeting, acquiring reply information of the target clarifying object to the demand problem in the clarifying meeting;
and carrying out voice conversion on the acquired reply information to obtain the reply voice.
8. A demand issue recovery device, comprising:
the semantic analysis unit is used for acquiring a demand problem sent by a user and performing semantic analysis on the demand problem to obtain semantic keywords;
the voice query unit is used for performing voice query according to the semantic keywords to obtain candidate clear voice and acquiring user information of the user;
the voice screening unit is used for screening the candidate clarified voice according to the user information and acquiring the voice information of the screened candidate clarified voice, wherein the voice information comprises voice recording time and voice label information;
the association score calculation unit is used for calculating an association score between the candidate clarified voice and the user according to the voice recording time and the voice tag information of the filtered candidate clarified voice, and the association score is used for representing the degree of association between the corresponding candidate clarified voice and the user;
and the question replying unit is used for replying the question to the demand question by the candidate clarified voice corresponding to the maximum association score.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 7 when executing the computer program.
10. A storage medium storing a computer program, characterized in that the computer program realizes the steps of the method according to any one of claims 1 to 7 when executed by a processor.
CN202011348081.6A 2020-11-26 2020-11-26 Demand question reply method and device, terminal equipment and storage medium Active CN112565663B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011348081.6A CN112565663B (en) 2020-11-26 2020-11-26 Demand question reply method and device, terminal equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011348081.6A CN112565663B (en) 2020-11-26 2020-11-26 Demand question reply method and device, terminal equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112565663A true CN112565663A (en) 2021-03-26
CN112565663B CN112565663B (en) 2022-11-18

Family

ID=75046843

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011348081.6A Active CN112565663B (en) 2020-11-26 2020-11-26 Demand question reply method and device, terminal equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112565663B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180096681A1 (en) * 2016-10-03 2018-04-05 Google Inc. Task initiation using long-tail voice commands
CN110659360A (en) * 2019-10-09 2020-01-07 初米网络科技(上海)有限公司 Man-machine conversation method, device and system
CN111046132A (en) * 2019-10-25 2020-04-21 众安信息技术服务有限公司 Customer service question and answer processing method and system for retrieving multiple rounds of conversations
CN111563182A (en) * 2020-04-28 2020-08-21 深圳震有科技股份有限公司 Voice conference record storage processing method and device
CN111797214A (en) * 2020-06-24 2020-10-20 深圳壹账通智能科技有限公司 FAQ database-based problem screening method and device, computer equipment and medium
CN111858877A (en) * 2020-06-17 2020-10-30 平安科技（深圳）有限公司 Multi-type question intelligent question answering method, system, equipment and readable storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180096681A1 (en) * 2016-10-03 2018-04-05 Google Inc. Task initiation using long-tail voice commands
CN110659360A (en) * 2019-10-09 2020-01-07 初米网络科技(上海)有限公司 Man-machine conversation method, device and system
CN111046132A (en) * 2019-10-25 2020-04-21 众安信息技术服务有限公司 Customer service question and answer processing method and system for retrieving multiple rounds of conversations
CN111563182A (en) * 2020-04-28 2020-08-21 深圳震有科技股份有限公司 Voice conference record storage processing method and device
CN111858877A (en) * 2020-06-17 2020-10-30 平安科技（深圳）有限公司 Multi-type question intelligent question answering method, system, equipment and readable storage medium
CN111797214A (en) * 2020-06-24 2020-10-20 深圳壹账通智能科技有限公司 FAQ database-based problem screening method and device, computer equipment and medium

Also Published As

Publication number Publication date
CN112565663B (en) 2022-11-18

Similar Documents

Publication Publication Date Title
US10777207B2 (en) Method and apparatus for verifying information
WO2022095380A1 (en) Ai-based virtual interaction model generation method and apparatus, computer device and storage medium
CN112765336B (en) Barrage management method and device, terminal equipment and storage medium
CN109636345B (en) Intelligent management method and system for business handling workflow
CN112925898B (en) Question-answering method and device based on artificial intelligence, server and storage medium
CN110895568B (en) Method and system for processing court trial records
CN109992781B (en) Text feature processing method and device and storage medium
CN109460503B (en) Answer input method, answer input device, storage medium and electronic equipment
CN112329586B (en) Customer return visit method and device based on emotion recognition and computer equipment
US11687796B2 (en) Document type-specific quality model
CN113032520A (en) Information analysis method and device, electronic equipment and computer readable storage medium
CN107729491B (en) Method, device and equipment for improving accuracy rate of question answer search
CN113094478A (en) Expression reply method, device, equipment and storage medium
CN117114475A (en) Comprehensive capability assessment system based on multidimensional talent assessment strategy
CN114860742A (en) Artificial intelligence-based AI customer service interaction method, device, equipment and medium
CN109885727A (en) Data pushing method, device, electronic equipment and system
CN113850077A (en) Topic identification method, device, server and medium based on artificial intelligence
CN112367494B (en) Online conference communication method and device based on AI and computer equipment
US20200020318A1 (en) Analysis of content sources for automatic generation of training content
CN112565663B (en) Demand question reply method and device, terminal equipment and storage medium
CN112085577A (en) On-line bidding question answering method, device, system and storage medium
CN114186048A (en) Question-answer replying method and device based on artificial intelligence, computer equipment and medium
CN114202363B (en) Artificial intelligence-based call method, device, computer equipment and medium
CN112786041B (en) Voice processing method and related equipment
US12045780B2 (en) Online interview method and system

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