CN109451188B - Method and device for differential self-help response, computer equipment and storage medium - Google Patents

Method and device for differential self-help response, computer equipment and storage medium Download PDF

Info

Publication number
CN109451188B
CN109451188B CN201811446908.XA CN201811446908A CN109451188B CN 109451188 B CN109451188 B CN 109451188B CN 201811446908 A CN201811446908 A CN 201811446908A CN 109451188 B CN109451188 B CN 109451188B
Authority
CN
China
Prior art keywords
current
client
information
broadcast
emotion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811446908.XA
Other languages
Chinese (zh)
Other versions
CN109451188A (en
Inventor
张垒
邢艳
邹芳
李晋
占敏敏
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 Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen 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 Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Priority to CN201811446908.XA priority Critical patent/CN109451188B/en
Publication of CN109451188A publication Critical patent/CN109451188A/en
Application granted granted Critical
Publication of CN109451188B publication Critical patent/CN109451188B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • 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
    • G10L17/00—Speaker identification or verification techniques
    • 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
    • G10L25/63—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 estimating an emotional state
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • H04M3/4936—Speech interaction details

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Physics & Mathematics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Child & Adolescent Psychology (AREA)
  • General Health & Medical Sciences (AREA)
  • Hospice & Palliative Care (AREA)
  • Psychiatry (AREA)
  • Computational Linguistics (AREA)
  • Telephonic Communication Services (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

The application provides a differential self-help response method, a differential self-help response device, computer equipment and a storage medium, and relates to the field of voice interaction, wherein the method comprises the following steps: acquiring current call information of a client in real time; analyzing the current call information to obtain current call content and current voiceprint information; inputting the current voiceprint information into an emotion database, and screening to obtain corresponding current client emotion characterization parameters; acquiring current broadcast content corresponding to sensitive words in current call content and current response voice corresponding to client emotion characterization parameters; and sending the current broadcast content and the current response voice to the client terminal so that the client broadcasts the current broadcast content by using the current response voice. According to the method and the device, voiceprint information and conversation content of the client are collected in real time in the conversation process, emotion information and conversation willingness of the client are recognized, broadcasting voice is dynamically adjusted according to the emotion of the client, corresponding broadcasting content is matched according to the current conversation content, and interaction integrity of the client is improved.

Description

Method and device for differential self-help response, computer equipment and storage medium
Technical Field
The present application relates to the field of voice interaction technologies, and in particular, to a method and an apparatus for differential self-help response, a computer device, and a storage medium.
Background
The self-service answering system can automatically answer most of the consultation information of the client according to the preset scene and the preset information, thereby greatly reducing the working pressure of customer service staff and being widely applied to the consultation work of various industries, such as telephone charge inquiry. However, in the interaction process between the existing self-service answering system and the client, only a single broadcast voice can be used, the volume, the speed, the emotion and the like of the existing self-service answering system cannot be adjusted in real time according to the actual interaction condition with the client, the experience feeling of the client when using the self-service answering system is very poor, the terminal conversation in advance is easily caused, and the consultation function cannot be realized.
Disclosure of Invention
The main purpose of the application is to provide a differential self-service answering method, a differential self-service answering device, a computer device and a storage medium, and the method and the device aim at solving the defects that the existing self-service answering system sets a rigid board in the interaction process and the interaction integrity is poor.
In order to achieve the above object, the present application provides a differential self-help answering method, which is applied to an answering terminal, and comprises:
acquiring current call information of a client in real time;
analyzing the current call information to obtain current call content and current voiceprint information;
inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information, wherein the emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters;
respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
and sending the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
Further, the step of inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain a current client emotion characterization parameter corresponding to the current voiceprint information includes:
analyzing the current voiceprint information to obtain current voiceprint parameters of the client, wherein the current voiceprint parameters comprise the speech speed, the tone and the volume of the client;
inputting the current voiceprint parameters into an emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall, wherein the emotion database is formed by correspondingly setting numerical value intervals of multiple groups of voiceprint parameters and client emotion characterization parameters.
Further, the step of respectively obtaining the current broadcast content corresponding to the current call content and the current response voice corresponding to the client emotion characterization parameter includes:
inputting the current call content into a broadcast database which is constructed in advance, and screening to obtain the current broadcast content;
and inputting the current client emotion characterization parameters into a pre-constructed response voice database, and screening to obtain the current response voice, wherein the broadcast database comprises a plurality of groups of corresponding call contents and broadcast contents, and the response voice database comprises a plurality of groups of corresponding client emotion characterization parameters and response voices.
Further, the step of inputting the current call content into a broadcast database constructed in advance and obtaining the current broadcast content through screening includes:
identifying sensitive words contained in the current call content;
inputting the sensitive words into the broadcast database, and screening to obtain broadcast contents corresponding to the sensitive words;
and setting the broadcast content corresponding to the sensitive words as the current broadcast content.
Further, before the step of collecting the current call information of the client in real time, the method comprises the following steps
Acquiring first call information, analyzing the first call information to obtain tone characteristics, and selecting a scene carried with the first call information;
inputting the tone features into a pre-constructed gender database, and screening to obtain the gender of the client corresponding to the tone features, wherein the gender database consists of a plurality of groups of tone features and the gender of the client;
inputting the customer gender into a pre-constructed broadcast voice library, screening to obtain broadcast voice corresponding to the customer gender, inputting scene selection into a pre-constructed scene database, screening to obtain scene content corresponding to the scene selection, wherein the broadcast voice library is formed by correspondingly inputting two groups of customer gender and broadcast voice, and the scene database is formed by correspondingly forming a plurality of groups of scene selections and scene contents;
and sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the first broadcast information to the client terminal, the method includes:
and binding the gender of the customer with the pre-entered personal information of the customer.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the first broadcast information to the client terminal, the method includes:
obtaining emotion change information of the client according to the change of the current client emotion characterization parameter in unit call time;
inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client character corresponding to the emotion change information;
and binding the current customer character with the customer personal information.
The application also provides a device that self-service answer of difference nature, includes:
the acquisition module is used for acquiring the current call information of the client in real time;
the first analysis module is used for analyzing the current call information to obtain current call content and current voiceprint information;
the first screening module is used for inputting the current voiceprint information into a pre-constructed emotion database and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information;
the acquisition module is used for respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
and the first sending module is used for sending the current broadcast content and the current response voice to the client terminal so that the client broadcasts the current broadcast content by using the current response voice.
Further, the first screening module includes:
the analysis unit is used for analyzing the current voiceprint information to obtain the current voiceprint parameters of the client;
and the first screening unit is used for inputting the current voiceprint parameters into the emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall.
Further, the obtaining module includes:
the second screening unit is used for inputting the current call content into a pre-constructed broadcast database and screening to obtain the current broadcast content;
and the third screening unit is used for inputting the current client emotion characterization parameters into a pre-constructed response voice database, screening to obtain the current response voice, and the broadcast database is correspondingly composed of a plurality of groups of call contents and broadcast contents.
Further, the second screening unit includes:
the identifying subunit is used for identifying sensitive words contained in the current call content;
the screening subunit is used for inputting the sensitive words into the broadcast database and screening to obtain broadcast contents corresponding to the sensitive words;
and the setting subunit is used for setting the broadcast content corresponding to the sensitive words as the current broadcast content.
Further, the apparatus further comprises:
the second analysis module is used for acquiring first call information and analyzing the first call information to obtain tone characteristics;
the second screening module is used for inputting the tone characteristics into a pre-constructed gender database and screening to obtain the gender of the client corresponding to the tone characteristics;
the third screening module is used for inputting the customer gender into a pre-constructed broadcast voice database, screening to obtain broadcast voice corresponding to the customer gender, inputting the scene selection into a pre-constructed scene database, and screening to obtain scene content corresponding to the scene selection;
and the second sending module is used for sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, the apparatus further comprises:
and the binding module is used for binding the gender of the customer with the pre-recorded personal information of the customer.
Further, the apparatus further comprises:
the third analysis module is used for obtaining the emotion change information of the client according to the change of the current client emotion representation parameter in unit conversation time;
the fourth screening module is used for inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client characters corresponding to the emotion change information;
and the second binding module is used for binding the current customer character with the customer personal information.
The present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method when executing the computer program.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method as set forth above.
According to the method, the device, the computer equipment and the storage medium for differential self-service response, the current voiceprint information and the current call content of the client are collected in real time in the call process, the current emotion information and the call will of the client are recognized, the speech speed, the tone and the volume of the broadcast speech are dynamically adjusted according to the emotion of the client, meanwhile, the broadcast content which is in line with the call will of the client is matched according to the current call content, and the interaction integrity of the broadcast content and the client is improved.
Drawings
FIG. 1 is a schematic diagram illustrating steps of a method for differentiated self-help response according to an embodiment of the present application;
FIG. 2 is a block diagram of a device for differential self-help response according to an embodiment of the present disclosure;
fig. 3 is a block diagram schematically illustrating a structure of a computer device according to an embodiment of the present application.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
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.
Referring to fig. 1, an embodiment of the present application provides a differential self-help answering method, including:
s1: acquiring current call information of a client in real time;
s2: analyzing the current call information to obtain current call content and current voiceprint information;
s3: inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information, wherein the emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters;
s4: respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
s5: and sending the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
In this embodiment, after the response terminal selects the first broadcast information, a communication channel is established with the client based on the selected scene flow, and the current call information of the client is collected in the whole course of the call process, which is described by taking a specific certain time as an example. The response terminal collects the current call information of the client in real time, analyzes the current call information according to different information types, and separates the current call content from the current voiceprint information in the current call information so as to facilitate the analysis and matching of the next step. The information type refers to a data type in the current call information, namely text data and voiceprint information. The current voiceprint information is obtained by directly extracting a voice signal of the current call information into an electric signal; the current call content is text data in the current call information, for example, two words of "query" are text data, and the voice data of two words of "query" is voiceprint information. The current voiceprint information collected by the answering terminal comprises the current speech speed, tone and volume of the client. And after the response terminal analyzes the current voiceprint information, inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain a current client emotion characterization parameter corresponding to the current voiceprint information. The emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters. A developer forms a corresponding numerical value interval by detecting numerical values of the speed, the tone and the volume of a person under different emotions and carrying out classified statistics on the detected numerical values, thereby training to form an emotion database. And the response terminal calculates the speech speed, the tone and the volume of the client through the current conversation information and generates respective corresponding specific numerical values. For example, the speech rate, pitch and volume correspond to specific values of 120, 60, 70. The response terminal can calculate the speech rate according to the number of the received complete current conversation information syllables divided by the time of the current conversation information; the tone and the volume can be obtained by directly measuring the frequency and the amplitude of the voice signal of the current call information, wherein the frequency of the voice signal corresponds to the tone, and the amplitude corresponds to the volume. When the emotion database is applied, the response terminal compares the specific numerical value of the current voiceprint information of the client with the emotion database, and the current emotion characterization parameters of the client can be identified by matching the corresponding numerical value type and interval. For example, the values of speech rate, pitch and volume corresponding to the "calm" state of the emotion characterization parameter in the emotion database are 120, 50-70 and 40-60; the numerical intervals of the speech speed, the tone and the volume corresponding to the restlessness state are 120-150, 70-80 and 60-80; the numerical intervals of the speech speed, the tone and the volume corresponding to the 'angry' state are 150-180, 80-90 and 80-100. The specific values of the current speech speed, the tone and the volume of the client are 120, 60 and 50, and the current emotion characterization parameter of the client can be determined to be 'calm' through comparison. And after the response terminal acquires the current emotion characterization parameters of the client, inputting the current call content into a pre-constructed broadcast database, and screening according to the sensitive words in the call information to obtain the current broadcast content. And simultaneously, inputting the current client emotion characterization parameters into a pre-constructed response voice library, and screening to obtain the current response voice. The broadcast database is composed of a plurality of groups of corresponding call contents and sensitive words of the broadcast contents, and the response voice database is composed of a plurality of groups of corresponding client emotion characterization parameters and response voices. For example, when the terminal recognizes that the client is impatient, the speech speed of the response speech obtained by screening is faster, the tone of the response speech is soft, and the volume of the response speech is larger. Meanwhile, the terminal identifies the sensitive words in the current call content, matches the sensitive words with the response voice library, and outputs the current broadcast content corresponding to the sensitive words. For example, if the terminal recognizes that the call content of the client contains 'personal insurance', the terminal outputs pre-recorded corresponding broadcast content, such as text information of the type and the allowance of the personal insurance, according to the sensitive word 'personal insurance'. And the response terminal sends the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
Further, the step of screening and obtaining the current client emotion characterization parameters corresponding to the current voiceprint information according to the current voiceprint information input into a pre-constructed emotion database comprises the following steps:
s301: analyzing the current voiceprint information to obtain current voiceprint parameters of the client, wherein the current voiceprint parameters comprise the speech speed, the tone and the volume of the client;
s302: inputting the current voiceprint parameters into an emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall, wherein the emotion database is formed by correspondingly setting numerical value intervals of multiple groups of voiceprint parameters and client emotion characterization parameters.
In this embodiment, the response terminal analyzes the collected current voiceprint information of the client, converts the current voiceprint information into specific voiceprint parameters, and obtains specific values of the speed, the pitch, and the volume of the current client. The speech rate can be obtained by calculating the number of the syllables in the collected current call information of the client according to the unit time. For example, if the number of the syllables in the current call information of the client collected by the answering terminal in 5S is 100, the speech rate of the current user is 120 words/S. The tone and the volume can be obtained by detecting a sound wave signal in the current call information, wherein the sound wave signal has a certain frequency and a certain frequency amplitude, the frequency corresponds to the tone, and the frequency amplitude corresponds to the volume. The answering terminal correspondingly expresses the speed, the tone and the volume of the analyzed current voiceprint information as specific current voiceprint parameters, for example, the speed of speech of the current user is 120, the tone is 60 and the volume is 70. An emotion database is built in the response terminal, and developers perform grouping entry generation on various emotions and the corresponding speech speed, tone and volume numerical value intervals according to the prior test. And the response terminal inputs the analyzed specific parameter values of the speech speed, the tone and the volume into an emotion database for screening, and obtains the corresponding current client emotion characterization parameters according to the numerical value intervals in which the parameter values respectively fall. For example, the value intervals of the speech rate, the pitch and the volume corresponding to the "restlessness" state in the emotion database are 150-.
Further, according to the step of respectively obtaining the current broadcast content corresponding to the current call content and the current response voice corresponding to the client emotion characterization parameter, the method comprises the following steps:
s401: inputting the current call content into a broadcast database which is constructed in advance, and screening to obtain the current broadcast content;
s402: and inputting the current client emotion characterization parameters into a pre-constructed response voice database, and screening to obtain the current response voice, wherein the broadcast database comprises a plurality of groups of corresponding call contents and broadcast contents, and the response voice database comprises a plurality of groups of corresponding client emotion characterization parameters and response voices.
In this embodiment, the response terminal inputs the current call content into a broadcast database that is constructed in advance. The broadcast database is formed by a plurality of groups of preset call contents which are input by developers in advance and corresponding broadcast contents. The response terminal identifies the current call content of the client through an ASR (voice recognition technology), converts the current call content into text information, identifies the sensitive words in the text information, and screens the corresponding current broadcast content from the broadcast database according to the identified sensitive words. For example, if the response terminal recognizes that the current call content of the client contains 'personal insurance', the response terminal screens the current broadcast content corresponding to the current call content, which is pre-recorded, from the broadcast database according to the sensitive word 'personal insurance', such as the type of personal insurance, the content of the premium and the like. Meanwhile, the response terminal inputs the current client emotion characterization parameters into a pre-constructed response voice library, and the current response voice is obtained through screening. The response voice library comprises a plurality of groups of response voices which are pre-recorded and correspond to different client emotion characterization parameters, and the response voices comprise different speeds, tone and volumes. For example, when the current emotion characterization parameter of the client is "calm", the corresponding response speech has moderate speech speed, gentle speech and moderate volume. And the response terminal inputs the client emotion characterization parameters into a response voice library and then performs screening to obtain response voice corresponding to the current client emotion characterization parameters.
Further, the step of inputting the current call content into a broadcast database constructed in advance and obtaining the current broadcast content through screening includes:
s4011: identifying sensitive words contained in the current call content;
s4012: inputting the sensitive words into the broadcast database, and screening to obtain broadcast contents corresponding to the sensitive words;
s4013: and setting the broadcast content corresponding to the sensitive words as the current broadcast content.
In this embodiment, the answering terminal recognizes the current call content of the client by ASR (speech recognition technology), and converts it into text information. At least one sensitive word preset by a developer is input into the response terminal and stored in the sensitive word bank. And the response terminal has different sensitive word banks according to different response scenes. For example, the sensitive word in the query scenario is different from the sensitive word in the return visit scenario, and the two sensitive words have two different sensitive word banks. When the response terminal identifies the sensitive words, the sensitive word bank corresponding to the response process needs to be selected according to the current response process. And the response terminal sequentially traverses each sensitive word in the sensitive word library and judges whether the sensitive word is contained in the text information, so that the input sensitive word contained in the text information is identified and obtained. After the response terminal identifies the sensitive words contained in the current call content, the sensitive words are input into the broadcast database, so that the broadcast content corresponding to the sensitive words is screened out, and the broadcast content corresponding to the screened sensitive words is set as the current broadcast content so as to perform the next action. For example, the response terminal inquires whether the client purchases the endowment insurance in the previous broadcast information sent to the client, and if the response terminal identifies that the call content fed back by the client is 'available', the response terminal calls the broadcast content corresponding to the call content, such as the type and the premium of the endowment insurance, which is recorded in advance from the broadcast database according to the 'available' sensitive word.
Further, before the step of collecting the current call information of the client in real time, the method comprises the following steps
S6: acquiring first call information, analyzing the first call information to obtain tone characteristics, and selecting a scene carried with the first call information;
s7: inputting the tone features into a pre-constructed gender database, and screening to obtain the gender of the client corresponding to the tone features, wherein the gender database consists of a plurality of groups of tone features and the gender of the client;
s8: inputting the customer gender into a pre-constructed broadcast voice library, screening to obtain broadcast voice corresponding to the customer gender, inputting scene selection into a pre-constructed scene database, screening to obtain scene content corresponding to the scene selection, wherein the broadcast voice library is formed by correspondingly inputting two groups of customer gender and broadcast voice, and the scene database is formed by correspondingly forming a plurality of groups of scene selections and scene contents;
s9: and sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
In this embodiment, when the answering terminal first calls the customer, the gender of the customer needs to be judged so as to adapt to the broadcast voice of the appropriate gender, and the conversation experience of the customer is improved. The response terminal converts the voice signal of the first call information into an electric signal by collecting the first call information when the current client makes the first call, directly extracts and obtains the voiceprint information, and analyzes and obtains the frequency corresponding to the voice signal, namely the tone characteristic corresponding to the client. The tone of the male voice is different from that of the female voice, the tone of the male voice is relatively low, and the tone of the female voice is relatively fine, namely, the frequency of the male voice is relatively low, and the frequency of the female voice is relatively high. The response terminal is pre-constructed with corresponding tone intervals according to different genders, so that the tone characteristics can be input into a pre-constructed gender database, and the gender of the client corresponding to the current client can be obtained by screening through comparing the tone characteristics with the tone intervals. The first call information carries a scene selection, for example, a scene corresponding to the user selection button 1 is selected as a consultation service scene in the call process. The response terminal is provided with different broadcast information according to different scenes. For example, the scene content of the collection scene is more resourceful in terms of words and has more severe voice; the scene content word of the return visit scene is milder, and the tone is softer. The instruction for selecting the scene can be sent by the response terminal or the client. When the answering terminal actively calls the client, the instruction of the scene selection is sent by the answering terminal. When the client calls the answering terminal, the instruction of the scene selection needs to be sent by the client, and the scene selection information is contained in the first-time call information. And the response terminal inputs the gender of the client into a pre-constructed broadcast voice database, and the broadcast voice corresponding to the gender of the client is obtained through screening. Wherein, report the pronunciation storehouse and correspond by two sets of customer sexes and report the pronunciation and constitute. For example, when the gender of the client is female, the second broadcast message is female voice. And simultaneously, inputting the scene selection into a pre-constructed scene database, and screening to obtain the scene content corresponding to the scene selection. And the response terminal sends the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s10: and binding the gender of the customer with the pre-recorded personal information of the customer.
In this embodiment, after the response terminal outputs the current broadcast information, the previously identified gender of the client may be bound to personal information of the client, such as a mobile phone number or a name of the client. The personal information of the client is collected in advance by a developer and then is input into the response terminal, meanwhile, the collected personal information of the client is incomplete, and no known information exists in the category of the gender of the client. After the customer gender and the personal information are bound, the response terminal can directly identify the customer gender through the personal information, such as a mobile phone number, when communicating with the customer again, so that the corresponding broadcast voice is directly matched without being identified again, the time is effectively saved, and the working efficiency is improved.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s11: obtaining emotion change information of the client according to the change of the current client emotion characterization parameter in unit call time;
s12: inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client character corresponding to the emotion change information;
s13: and binding the current customer character with the customer personal information.
In this embodiment, the responding terminal constantly records the emotion change information of the client during the communication with the client, and obtains the emotion change information of the client according to the change of the current emotion characterization parameter of the client in the unit communication time. The emotion change information comprises the call time of the change of the emotion characteristic parameters and specific emotion characteristic parameters before and after the change of the emotion characteristic parameters. A character information base is pre-constructed in the response terminal, and various character information and emotion representation parameter changes and corresponding call time intervals during the change are correspondingly grouped through pre-testing. For example, the personality information of the client is "impatience", the corresponding emotion characterization parameter of the client is initially "calm", the subsequent change is "impatience", and the call time for generating the change is 1-3 minutes. After a call is finished, the answering terminal inputs the personality information base for screening according to the change of emotion characterization parameters of a client in the whole call process and the corresponding call time when the emotion characterization parameters change, so that corresponding personality information of the client in the call, namely the current client personality, is obtained, and the current client personality is bound with personal information of the client, such as a mobile phone number, so that when the call is made with the client again next time, appropriate initial broadcast voice is directly selected according to the current client personality, such as the impatient client selects the initial broadcast voice with the higher voice speed, and the client experience is effectively improved.
According to the method for differential self-help response, the current voiceprint information and the current call content of the client are collected in real time in the call process, the current emotion information and the call will of the client are recognized, the speed, the tone and the volume of the broadcast voice are dynamically adjusted according to the emotion of the client, meanwhile, the broadcast content which is in line with the call will of the client is matched according to the current call content, and the interaction integrity with the client is improved.
Referring to fig. 2, an embodiment of the present application further provides a device for differential self-help response, including:
the acquisition module 1 is used for acquiring the current call information of a client in real time;
the first analysis module 2 is used for analyzing the current call information to obtain current call content and current voiceprint information;
the first screening module 3 is used for inputting the current voiceprint information into a pre-constructed emotion database and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information;
the obtaining module 4 is configured to obtain current broadcast content corresponding to a sensitive word in the current call content and current response voice corresponding to the client emotion characterization parameter respectively;
and a first sending module 5, configured to send the current broadcast content and the current response voice to the client terminal, so that the client broadcasts the current broadcast content using the current response voice.
In this embodiment, after the response terminal selects the first broadcast information, a communication channel is established with the client based on the selected scene flow, and the current call information of the client is collected in the whole course of the call process, which is described by taking a specific certain time as an example. The response terminal collects the current call information of the client in real time, analyzes the current call information according to different information types, and separates the current call content from the current voiceprint information in the current call information so as to facilitate the analysis and matching of the next step. The information type refers to a data type in the current call information, namely text data and voiceprint information. The current voiceprint information is obtained by directly extracting a voice signal of the current call information into an electric signal; the current call content is text data in the current call information, for example, two words of "query" are text data, and the voice data of two words of "query" is voiceprint information. The current voiceprint information collected by the answering terminal comprises the current speech speed, tone and volume of the client. And after the response terminal analyzes the current voiceprint information, inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain a current client emotion characterization parameter corresponding to the current voiceprint information. The emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters. A developer forms a corresponding numerical value interval by detecting numerical values of the speed, the tone and the volume of a person under different emotions and carrying out classified statistics on the detected numerical values, thereby training to form an emotion database. And the response terminal calculates the speech speed, the tone and the volume of the client through the current conversation information and generates respective corresponding specific numerical values. For example, the speech rate, pitch and volume correspond to specific values of 120, 60, 70. The response terminal can calculate the speech rate according to the number of the received complete current conversation information syllables divided by the time of the current conversation information; the tone and the volume can be obtained by directly measuring the frequency and the amplitude of the voice signal of the current call information, wherein the frequency of the voice signal corresponds to the tone, and the amplitude corresponds to the volume. When the emotion database is applied, the response terminal compares the specific numerical value of the current voiceprint information of the client with the emotion database, and the current emotion characterization parameters of the client can be identified by matching the corresponding numerical value type and interval. For example, the values of speech rate, pitch and volume corresponding to the "calm" state of the emotion characterization parameter in the emotion database are 120, 50-70 and 40-60; the numerical intervals of the speech speed, the tone and the volume corresponding to the restlessness state are 120-150, 70-80 and 60-80; the numerical intervals of the speech speed, the tone and the volume corresponding to the 'angry' state are 150-180, 80-90 and 80-100. The specific values of the current speech speed, the tone and the volume of the client are 120, 60 and 50, and the current emotion characterization parameter of the client can be determined to be 'calm' through comparison. And after the response terminal acquires the current emotion characterization parameters of the client, inputting the current call content into a pre-constructed broadcast database, and screening according to the sensitive words in the call information to obtain the current broadcast content. And simultaneously, inputting the current client emotion characterization parameters into a pre-constructed response voice library, and screening to obtain the current response voice. The broadcast database is composed of a plurality of groups of corresponding call contents and sensitive words of the broadcast contents, and the response voice database is composed of a plurality of groups of corresponding client emotion characterization parameters and response voices. For example, when the terminal recognizes that the client is impatient, the speech speed of the response speech obtained by screening is faster, the tone of the response speech is soft, and the volume of the response speech is larger. Meanwhile, the terminal identifies the sensitive words in the current call content, matches the sensitive words with the response voice library, and outputs the current broadcast content corresponding to the sensitive words. For example, if the terminal recognizes that the call content of the client contains 'personal insurance', the terminal outputs pre-recorded corresponding broadcast content, such as text information of the type and the allowance of the personal insurance, according to the sensitive word 'personal insurance'. And the response terminal sends the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
Further, the first screening module 3 includes:
the analysis unit is used for analyzing the current voiceprint information to obtain the current voiceprint parameters of the client;
and the first screening unit is used for inputting the current voiceprint parameters into the emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall.
In this embodiment, the response terminal analyzes the collected current voiceprint information of the client, converts the current voiceprint information into specific voiceprint parameters, and obtains specific values of the speed, the pitch, and the volume of the current client. The speech rate can be obtained by calculating the number of the syllables in the collected current call information of the client according to the unit time. For example, if the number of the syllables in the current call information of the client collected by the answering terminal in 5S is 100, the speech rate of the current user is 120 words/S. The tone and the volume can be obtained by detecting a sound wave signal in the current call information, wherein the sound wave signal has a certain frequency and a certain frequency amplitude, the frequency corresponds to the tone, and the frequency amplitude corresponds to the volume. The answering terminal correspondingly expresses the speed, the tone and the volume of the analyzed current voiceprint information as specific current voiceprint parameters, for example, the speed of speech of the current user is 120, the tone is 60 and the volume is 70. An emotion database is built in the response terminal, and developers perform grouping entry generation on various emotions and the corresponding speech speed, tone and volume numerical value intervals according to the prior test. And the response terminal inputs the analyzed specific parameter values of the speech speed, the tone and the volume into an emotion database for screening, and obtains the corresponding current client emotion characterization parameters according to the numerical value intervals in which the parameter values respectively fall. For example, the value intervals of the speech rate, the pitch and the volume corresponding to the "restlessness" state in the emotion database are 150-.
Further, the obtaining module 4 includes:
the second screening unit is used for inputting the current call content into a pre-constructed broadcast database and screening to obtain the current broadcast content;
and the third screening unit is used for inputting the current client emotion characterization parameters into a pre-constructed response voice database, screening to obtain the current response voice, and the broadcast database is correspondingly composed of a plurality of groups of call contents and broadcast contents.
In this embodiment, the response terminal inputs the current call content into a broadcast database that is constructed in advance. The broadcast database is formed by a plurality of groups of preset call contents which are input by developers in advance and corresponding broadcast contents. The response terminal identifies the current call content of the client through an ASR (voice recognition technology), converts the current call content into text information, identifies the sensitive words in the text information, and screens the corresponding current broadcast content from the broadcast database according to the identified sensitive words. For example, if the answering terminal identifies that the current call content of the client contains 'personal insurance', the current broadcast content corresponding to the current call content, which is pre-recorded, is obtained by screening from the broadcast database according to the sensitive word 'personal insurance', such as the type of personal insurance, the content of the premium and the like. Meanwhile, the response terminal inputs the current client emotion characterization parameters into a pre-constructed response voice library, and the current response voice is obtained through screening. The response voice library comprises a plurality of groups of response voices which are recorded in advance and correspond to different emotion information, and the response voices comprise different speeds, tone and volumes. For example, when the current emotion characterization parameter of the client is "calm", the corresponding response speech has moderate speech speed, gentle speech and moderate volume. And the response terminal inputs the client emotion characterization parameters into a response voice library and then performs screening to obtain response voice corresponding to the current client emotion characterization parameters.
Further, the second screening unit includes:
the identifying subunit is used for identifying sensitive words contained in the current call content;
the screening subunit is used for inputting the sensitive words into the broadcast database and screening to obtain broadcast contents corresponding to the sensitive words;
and the setting subunit is used for setting the broadcast content corresponding to the sensitive words as the current broadcast content.
In this embodiment, the answering terminal recognizes the current call content of the client by ASR (speech recognition technology), and converts it into text information. At least one sensitive word preset by a developer is input into the response terminal and stored in the sensitive word bank. And the response terminal has different sensitive word banks according to different response scenes. For example, the sensitive word in the query scenario is different from the sensitive word in the return visit scenario, and the two sensitive words have two different sensitive word banks. When the response terminal identifies the sensitive words, the sensitive word bank corresponding to the response process needs to be selected according to the current response process. And the response terminal sequentially traverses each sensitive word in the sensitive word library and judges whether the sensitive word is contained in the text information, so that the input sensitive word contained in the text information is identified and obtained. After the response terminal identifies the sensitive words contained in the current call content, the sensitive words are input into the broadcast database, so that the broadcast content corresponding to the sensitive words is screened out, and the broadcast content corresponding to the screened sensitive words is set as the current broadcast content so as to perform the next action. For example, the response terminal inquires whether the client purchases the endowment insurance in the previous broadcast information sent to the client, and if the response terminal identifies that the call content fed back by the client is 'available', the response terminal calls the broadcast content corresponding to the call content, such as the type and the premium of the endowment insurance, which is recorded in advance from the broadcast database according to the 'available' sensitive word.
Further, the processing device further comprises:
the second analysis module is used for acquiring first call information and analyzing the first call information to obtain tone characteristics;
the second screening module is used for inputting the tone characteristics into a pre-constructed gender database and screening to obtain the gender of the client corresponding to the tone characteristics;
the third screening module is used for inputting the customer gender into a pre-constructed broadcast voice database, screening to obtain broadcast voice corresponding to the customer gender, inputting the scene selection into a pre-constructed scene database, and screening to obtain scene content corresponding to the scene selection;
and the second sending module is used for sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
In this embodiment, when the answering terminal first calls the customer, the gender of the customer needs to be judged so as to adapt to the broadcast voice of the appropriate gender, and the conversation experience of the customer is improved. The response terminal converts the voice signal of the first call information into an electric signal by collecting the first call information when the current client makes the first call, directly extracts and obtains the voiceprint information, and analyzes and obtains the frequency corresponding to the voice signal, namely the tone characteristic corresponding to the client. The tone of the male voice is different from that of the female voice, the tone of the male voice is relatively low, and the tone of the female voice is relatively fine, namely, the frequency of the male voice is relatively low, and the frequency of the female voice is relatively high. The response terminal is pre-constructed with corresponding tone intervals according to different genders, so that the tone characteristics can be input into a pre-constructed gender database, and the gender of the client corresponding to the current client can be obtained by screening through comparing the tone characteristics with the tone intervals. The first call information carries a scene selection, for example, a scene corresponding to the user selection button 1 is selected as a consultation service scene in the call process. The response terminal is provided with different broadcast information according to different scenes. For example, the scene content of the collection scene is more resourceful in terms of words and has more severe voice; the scene content word of the return visit scene is milder, and the tone is softer. The instruction for selecting the scene can be sent by the response terminal or the client. When the answering terminal actively calls the client, the instruction of the scene selection is sent by the answering terminal. When the client calls the answering terminal, the instruction of the scene selection needs to be sent by the client, and the scene selection information is contained in the first-time call information. And the response terminal inputs the gender of the client into a pre-constructed broadcast voice database, and the broadcast voice corresponding to the gender of the client is obtained through screening. Wherein, report the pronunciation storehouse and correspond by two sets of customer sexes and report the pronunciation and constitute. For example, when the gender of the client is female, the second broadcast message is female voice. And simultaneously, inputting the scene selection into a pre-constructed scene database, and screening to obtain the scene content corresponding to the scene selection. And the response terminal sends the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, the processing device further comprises:
and the binding module is used for binding the gender of the customer with the pre-recorded personal information of the customer.
In this embodiment, after the response terminal outputs the current broadcast information, the previously identified gender of the client may be bound to personal information of the client, such as a mobile phone number or a name of the client. The personal information of the client is collected in advance by a developer and then is input into the response terminal, meanwhile, the collected personal information of the client is incomplete, and no known information exists in the category of the gender of the client. After the customer gender and the personal information are bound, the response terminal can directly identify the customer gender through the personal information, such as a mobile phone number, when communicating with the customer again, so that the corresponding broadcast voice is directly matched without being identified again, the time is effectively saved, and the working efficiency is improved.
Further, the processing device further comprises:
the third analysis module is used for obtaining the emotion change information of the client according to the change of the current client emotion representation parameter in unit conversation time;
the fourth screening module is used for inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client characters corresponding to the emotion change information;
and the second binding module is used for binding the current customer character with the customer personal information.
In this embodiment, the responding terminal constantly records the emotion change information of the client during the communication with the client, and obtains the emotion change information of the client according to the change of the current emotion characterization parameter of the client in the unit communication time. The emotion change information comprises the call time of the change of the emotion characteristic parameters and specific emotion characteristic parameters before and after the change of the emotion characteristic parameters. A character information base is pre-constructed in the response terminal, and various character information and emotion representation parameter changes and corresponding call time intervals during the change are correspondingly grouped through pre-testing. For example, the personality information of the client is "impatience", the corresponding emotion characterization parameter of the client is initially "calm", the subsequent change is "impatience", and the call time for generating the change is 1-3 minutes. After a call is finished, the answering terminal inputs the personality information base for screening according to the change of emotion characterization parameters of a client in the whole call process and the corresponding call time when the emotion characterization parameters change, so that corresponding personality information of the client in the call, namely the current client personality, is obtained, and the current client personality is bound with personal information of the client, such as a mobile phone number, so that when the call is made with the client again next time, appropriate initial broadcast voice is directly selected according to the current client personality, such as the impatient client selects the initial broadcast voice with the higher voice speed, and the client experience is effectively improved.
The device for differential self-help response provided in the embodiment identifies current emotion information and call willingness of a client by acquiring current voiceprint information and current call content of the client in real time in a call process, dynamically adjusts the speed, tone and volume of broadcast voice according to the emotion of the client, matches the broadcast content according with the call willingness of the client according to the current call content, and improves the interaction integrity with the client.
Referring to fig. 3, a computer device, which may be a server and whose internal structure may be as shown in fig. 3, is also provided in the embodiment of the present application. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data such as a mood database. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of differential self-help answering.
The processor executes the steps of the method for differential self-help response:
s1: acquiring current call information of a client in real time;
s2: analyzing the current call information to obtain current call content and current voiceprint information;
s3: inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information, wherein the emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters;
s4: respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
s5: and sending the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
Further, the step of screening and obtaining the current client emotion characterization parameters corresponding to the current voiceprint information according to the current voiceprint information input into a pre-constructed emotion database comprises the following steps:
s301: analyzing the current voiceprint information to obtain current voiceprint parameters of the client, wherein the current voiceprint parameters comprise the speech speed, the tone and the volume of the client;
s302: inputting the current voiceprint parameters into an emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall, wherein the emotion database is formed by correspondingly setting numerical value intervals of multiple groups of voiceprint parameters and client emotion characterization parameters.
Further, according to the step of respectively obtaining the current broadcast content corresponding to the current call content and the current response voice corresponding to the client emotion characterization parameter, the method comprises the following steps:
s401: inputting the current call content into a broadcast database which is constructed in advance, and screening to obtain the current broadcast content;
s402: and inputting the current client emotion characterization parameters into a pre-constructed response voice database, and screening to obtain the current response voice, wherein the broadcast database comprises a plurality of groups of corresponding call contents and broadcast contents, and the response voice database comprises a plurality of groups of corresponding client emotion characterization parameters and response voices.
Further, the step of inputting the current call content into a broadcast database constructed in advance and obtaining the current broadcast content through screening includes:
s4011: identifying sensitive words contained in the current call content;
s4012: inputting the sensitive words into the broadcast database, and screening to obtain broadcast contents corresponding to the sensitive words;
s4013: and setting the broadcast content corresponding to the sensitive words as the current broadcast content.
Further, before the step of collecting the current call information of the client in real time, the method comprises the following steps
S6: acquiring first call information, analyzing the first call information to obtain tone characteristics, and selecting a scene carried with the first call information;
s7: inputting the tone features into a pre-constructed gender database, and screening to obtain the gender of the client corresponding to the tone features, wherein the gender database consists of a plurality of groups of tone features and the gender of the client;
s8: inputting the customer gender into a pre-constructed broadcast voice library, screening to obtain broadcast voice corresponding to the customer gender, inputting scene selection into a pre-constructed scene database, screening to obtain scene content corresponding to the scene selection, wherein the broadcast voice library is formed by correspondingly inputting two groups of customer gender and broadcast voice, and the scene database is formed by correspondingly forming a plurality of groups of scene selections and scene contents;
s9: and sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s10: and binding the gender of the customer with the pre-recorded personal information of the customer.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s11: obtaining emotion change information of the client according to the change of the current client emotion characterization parameter in unit call time;
s12: inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client character corresponding to the emotion change information;
s13: and binding the current customer character with the customer personal information.
Those skilled in the art will appreciate that the architecture shown in fig. 3 is only a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects may be applied.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a method for differential self-help response, and specifically includes:
s1: acquiring current call information of a client in real time;
s2: analyzing the current call information to obtain current call content and current voiceprint information;
s3: inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information, wherein the emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters;
s4: respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
s5: and sending the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice.
Further, the step of screening and obtaining the current client emotion characterization parameters corresponding to the current voiceprint information according to the current voiceprint information input into a pre-constructed emotion database comprises the following steps:
s301: analyzing the current voiceprint information to obtain current voiceprint parameters of the client, wherein the current voiceprint parameters comprise the speech speed, the tone and the volume of the client;
s302: inputting the current voiceprint parameters into an emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall, wherein the emotion database is formed by correspondingly setting numerical value intervals of multiple groups of voiceprint parameters and client emotion characterization parameters.
Further, according to the step of respectively obtaining the current broadcast content corresponding to the current call content and the current response voice corresponding to the client emotion characterization parameter, the method comprises the following steps:
s401: inputting the current call content into a broadcast database which is constructed in advance, and screening to obtain the current broadcast content;
s402: and inputting the current client emotion characterization parameters into a pre-constructed response voice database, and screening to obtain the current response voice, wherein the broadcast database comprises a plurality of groups of corresponding call contents and broadcast contents, and the response voice database comprises a plurality of groups of corresponding client emotion characterization parameters and response voices.
Further, the step of inputting the current call content into a broadcast database constructed in advance and obtaining the current broadcast content through screening includes:
s4011: identifying sensitive words contained in the current call content;
s4012: inputting the sensitive words into the broadcast database, and screening to obtain broadcast contents corresponding to the sensitive words;
s4013: and setting the broadcast content corresponding to the sensitive words as the current broadcast content.
Further, before the step of collecting the current call information of the client in real time, the method comprises the following steps
S6: acquiring first call information, analyzing the first call information to obtain tone characteristics, and selecting a scene carried with the first call information;
s7: inputting the tone features into a pre-constructed gender database, and screening to obtain the gender of the client corresponding to the tone features, wherein the gender database consists of a plurality of groups of tone features and the gender of the client;
s8: inputting the customer gender into a pre-constructed broadcast voice library, screening to obtain broadcast voice corresponding to the customer gender, inputting scene selection into a pre-constructed scene database, screening to obtain scene content corresponding to the scene selection, wherein the broadcast voice library is formed by correspondingly inputting two groups of customer gender and broadcast voice, and the scene database is formed by correspondingly forming a plurality of groups of scene selections and scene contents;
s9: and sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s10: and binding the gender of the customer with the pre-recorded personal information of the customer.
Further, after the step of generating and using the broadcast voice broadcast the first broadcast information of the scene content and outputting the same to the client terminal, the method includes:
s11: obtaining emotion change information of the client according to the change of the current client emotion characterization parameter in unit call time;
s12: inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client character corresponding to the emotion change information;
s13: and binding the current customer character with the customer personal information.
In summary, the present application provides a method, an apparatus, a computer device, and a storage medium for differential self-help response.
It will be understood by those skilled in the art that all or part of the processes of the methods of the above embodiments may be implemented by hardware associated with instructions of a computer program, which may be stored on a non-volatile computer-readable storage medium, and when executed, may include processes of the above embodiments of the methods. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only for the preferred embodiment of the present application and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are intended to be included within the scope of the present application.

Claims (9)

1. A differential self-help answering method is applied to an answering terminal and comprises the following steps:
acquiring current call information of a client in real time;
analyzing the current call information to obtain current call content and current voiceprint information;
inputting the current voiceprint information into a pre-constructed emotion database, and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information, wherein the emotion database is formed by corresponding multiple groups of voiceprint information and client emotion characterization parameters;
respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
sending the current broadcast content and the current response voice to the client terminal so that the client terminal broadcasts the current broadcast content by using the current response voice;
obtaining emotion change information of the client according to the change of the current client emotion characterization parameter in unit call time;
inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client character corresponding to the emotion change information;
and binding the current customer character with the customer personal information, and directly selecting corresponding initial broadcast voice according to the current customer character when a customer corresponding to the customer personal information calls again next time.
2. The method of differential self-help response according to claim 1, wherein the step of inputting the current voiceprint information into a pre-established emotion database and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information comprises:
analyzing the current voiceprint information to obtain the current voiceprint parameters of the client, wherein the current voiceprint parameters comprise the speech speed, the tone and the volume of the client;
inputting the current voiceprint parameters into an emotion database, and screening out current client emotion characterization parameters corresponding to the current voiceprint parameters according to numerical value intervals in which the speech speed, the tone and the volume respectively fall, wherein the emotion database is formed by corresponding numerical value intervals of multiple groups of voiceprint parameters and client emotions.
3. The method for differential self-help response according to claim 1, wherein the step of respectively obtaining the current announced content corresponding to the current conversation content and the current response voice corresponding to the client emotion characterization parameter comprises:
inputting the current call content into a broadcast database which is constructed in advance, and screening to obtain the current broadcast content;
and inputting the current client emotion characterization parameters into a pre-constructed response voice database, and screening to obtain the current response voice, wherein the broadcast database comprises a plurality of groups of corresponding call contents and broadcast contents, and the response voice database comprises a plurality of groups of corresponding client emotion characterization parameters and response voices.
4. The method for differential self-help response according to claim 3, wherein the step of inputting the current call content into a pre-constructed broadcast database and screening the current broadcast content comprises:
identifying sensitive words contained in the current call content;
inputting the sensitive words into the broadcast database, and screening to obtain broadcast contents corresponding to the sensitive words;
and setting the broadcast content corresponding to the sensitive words as the current broadcast content.
5. The method of differential self-help response according to claim 1, wherein the step of collecting the current call information of the customer in real time is preceded by the step of collecting the current call information of the customer in real time, and comprises
Acquiring first call information, analyzing the first call information to obtain tone characteristics, and selecting a scene carried with the first call information;
inputting the tone features into a pre-constructed gender database, and screening to obtain the gender of the client corresponding to the tone features, wherein the gender database consists of a plurality of groups of tone features and the gender of the client;
inputting the customer gender into a pre-constructed broadcast voice library, screening to obtain broadcast voice corresponding to the customer gender, inputting scene selection into a pre-constructed scene database, screening to obtain scene content corresponding to the scene selection, wherein the broadcast voice library is formed by correspondingly inputting two groups of customer gender and broadcast voice, and the scene database is formed by correspondingly forming a plurality of groups of scene selections and scene contents;
and sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content by using the broadcast voice.
6. The method for differential self-help response according to claim 5, wherein the step of sending the scene content and the broadcast voice to the client terminal so that the client terminal broadcasts the scene content using the broadcast voice comprises:
and binding the gender of the customer with the pre-entered personal information of the customer.
7. A device for differentiated self-help response, comprising:
the acquisition module is used for acquiring the current call information of the client in real time;
the first analysis module is used for analyzing the current call information to obtain current call content and current voiceprint information;
the first screening module is used for inputting the current voiceprint information into a pre-constructed emotion database and screening to obtain current client emotion characterization parameters corresponding to the current voiceprint information;
the acquisition module is used for respectively acquiring current broadcast content corresponding to sensitive words in the current call content and current response voice corresponding to the client emotion characterization parameters;
a first sending module, configured to send the current broadcast content and the current response voice to the client terminal, so that the client terminal broadcasts the current broadcast content using the current response voice;
the third analysis module is used for obtaining the emotion change information of the client according to the change of the current client emotion representation parameter in unit conversation time;
the fourth screening module is used for inputting the emotion change information into a character database which is constructed in advance, and screening to obtain the current client characters corresponding to the emotion change information;
and the second binding module is used for binding the current customer character with the customer personal information, and directly selecting the corresponding initial broadcast voice according to the current customer character when the customer corresponding to the customer personal information calls again next time.
8. A computer device comprising a memory and a processor, the memory having stored therein a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method according to any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
CN201811446908.XA 2018-11-29 2018-11-29 Method and device for differential self-help response, computer equipment and storage medium Active CN109451188B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811446908.XA CN109451188B (en) 2018-11-29 2018-11-29 Method and device for differential self-help response, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811446908.XA CN109451188B (en) 2018-11-29 2018-11-29 Method and device for differential self-help response, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN109451188A CN109451188A (en) 2019-03-08
CN109451188B true CN109451188B (en) 2022-03-18

Family

ID=65555181

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811446908.XA Active CN109451188B (en) 2018-11-29 2018-11-29 Method and device for differential self-help response, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN109451188B (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110062117B (en) * 2019-04-08 2021-01-08 商客通尚景科技（上海）股份有限公司 Sound wave detection and early warning method
CN110265062A (en) * 2019-06-13 2019-09-20 上海指旺信息科技有限公司 Collection method and device after intelligence based on mood detection is borrowed
CN110719362A (en) * 2019-09-10 2020-01-21 北京百度网讯科技有限公司 Call processing method and device, electronic equipment and storage medium
CN111193834B (en) * 2019-12-16 2022-04-15 北京淇瑀信息科技有限公司 Man-machine interaction method and device based on user sound characteristic analysis and electronic equipment
CN111614845A (en) * 2020-04-21 2020-09-01 深圳追一科技有限公司 Human-computer dialogue human-set matching method and device, computer equipment and storage medium
CN111681681A (en) * 2020-05-22 2020-09-18 深圳壹账通智能科技有限公司 Voice emotion recognition method and device, electronic equipment and storage medium
CN111933138B (en) * 2020-08-20 2022-10-21 Oppo（重庆）智能科技有限公司 Voice control method, device, terminal and storage medium
CN112233699B (en) * 2020-10-13 2023-04-28 中移（杭州）信息技术有限公司 Voice broadcasting method, intelligent voice equipment and computer readable storage medium
TWI741937B (en) * 2021-01-20 2021-10-01 橋良股份有限公司 Judgment system for suitability of talents and implementation method thereof
TWI738610B (en) * 2021-01-20 2021-09-01 橋良股份有限公司 Recommended financial product and risk control system and implementation method thereof
CN113208592B (en) * 2021-03-29 2022-08-16 济南大学 Psychological test system with multiple answering modes
CN113643684B (en) * 2021-07-21 2024-02-27 广东电力信息科技有限公司 Speech synthesis method, device, electronic equipment and storage medium
CN113676527A (en) * 2021-08-10 2021-11-19 未鲲(上海)科技服务有限公司 Information pushing method, device, equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106874265A (en) * 2015-12-10 2017-06-20 深圳新创客电子科技有限公司 A kind of content outputting method matched with user emotion, electronic equipment and server
CN107293310A (en) * 2017-06-28 2017-10-24 上海航动科技有限公司 A kind of user emotion analysis method and system
CN108648768A (en) * 2018-04-16 2018-10-12 广州市菲玛尔咨询服务有限公司 A kind of consulting recommendation method and its management system

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103634472B (en) * 2013-12-06 2016-11-23 惠州Tcl移动通信有限公司 User mood and the method for personality, system and mobile phone is judged according to call voice
EP2887627A1 (en) * 2013-12-18 2015-06-24 Telefonica Digital España, S.L.U. Method and system for extracting out characteristics of a communication between at least one client and at least one support agent and computer program product thereof
CN106649404B (en) * 2015-11-04 2019-12-27 陈包容 Method and device for creating session scene database
CN107645523A (en) * 2016-07-21 2018-01-30 北京快乐智慧科技有限责任公司 A kind of method and system of mood interaction
CN106294774A (en) * 2016-08-11 2017-01-04 北京光年无限科技有限公司 User individual data processing method based on dialogue service and device
CN106776936B (en) * 2016-12-01 2020-02-18 上海智臻智能网络科技股份有限公司 Intelligent interaction method and system
CN108363706B (en) * 2017-01-25 2023-07-18 北京搜狗科技发展有限公司 Method and device for man-machine dialogue interaction
CN108735232A (en) * 2017-04-24 2018-11-02 北京理工大学 A kind of personality recognition methods and device
CN107329996B (en) * 2017-06-08 2021-06-29 三峡大学 Chat robot system and chat method based on fuzzy neural network
CN107301213A (en) * 2017-06-09 2017-10-27 腾讯科技（深圳）有限公司 Intelligent answer method and device
CN108197115B (en) * 2018-01-26 2022-04-22 上海智臻智能网络科技股份有限公司 Intelligent interaction method and device, computer equipment and computer readable storage medium

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106874265A (en) * 2015-12-10 2017-06-20 深圳新创客电子科技有限公司 A kind of content outputting method matched with user emotion, electronic equipment and server
CN107293310A (en) * 2017-06-28 2017-10-24 上海航动科技有限公司 A kind of user emotion analysis method and system
CN108648768A (en) * 2018-04-16 2018-10-12 广州市菲玛尔咨询服务有限公司 A kind of consulting recommendation method and its management system

Also Published As

Publication number Publication date
CN109451188A (en) 2019-03-08

Similar Documents

Publication Publication Date Title
CN109451188B (en) Method and device for differential self-help response, computer equipment and storage medium
CN111128223B (en) Text information-based auxiliary speaker separation method and related device
CN111311327A (en) Service evaluation method, device, equipment and storage medium based on artificial intelligence
CN112037799B (en) Voice interrupt processing method and device, computer equipment and storage medium
CN110246507B (en) Voice recognition method and device
CN110177182B (en) Sensitive data processing method and device, computer equipment and storage medium
CN110110038B (en) Telephone traffic prediction method, device, server and storage medium
CN109272993A (en) Recognition methods, device, computer equipment and the storage medium of voice class
CN111105782A (en) Session interaction processing method and device, computer equipment and storage medium
CN110769425B (en) Method and device for judging abnormal call object, computer equipment and storage medium
CN110990685A (en) Voice search method, voice search device, voice search storage medium and voice search device based on voiceprint
CN111858892A (en) Voice interaction method, device, equipment and medium based on knowledge graph
CN111312286A (en) Age identification method, age identification device, age identification equipment and computer readable storage medium
CN111223476A (en) Method and device for extracting voice feature vector, computer equipment and storage medium
CN113709313B (en) Intelligent quality inspection method, device, equipment and medium for customer service call data
KR100463706B1 (en) A system and a method for analyzing human emotion based on voice recognition through wire or wireless network
CN108717851B (en) Voice recognition method and device
CN108986786B (en) Voice interaction equipment rating method, system, computer equipment and storage medium
CN113194210B (en) Voice call access method and device
CN115050372A (en) Audio segment clustering method and device, electronic equipment and medium
CN115019802A (en) Speech intention recognition method and device, computer equipment and storage medium
CN114446325A (en) Information pushing method and device based on emotion recognition, computer equipment and medium
CN110931020B (en) Voice detection method and device
CN113593580A (en) Voiceprint recognition method and device
CN112712793A (en) ASR (error correction) method based on pre-training model under voice interaction and related equipment

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