CN108989589A - A kind of intelligent call semantics recognition communication method and intelligent conversational system - Google Patents
A kind of intelligent call semantics recognition communication method and intelligent conversational system Download PDFInfo
- Publication number
- CN108989589A CN108989589A CN201810794548.6A CN201810794548A CN108989589A CN 108989589 A CN108989589 A CN 108989589A CN 201810794548 A CN201810794548 A CN 201810794548A CN 108989589 A CN108989589 A CN 108989589A
- Authority
- CN
- China
- Prior art keywords
- call
- cloud server
- verbal system
- context
- answer
- 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.)
- Pending
Links
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
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/527—Centralised call answering arrangements not requiring operator intervention
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The present invention provides a kind of intelligent call semantics recognition communication method and intelligent conversational systems.The intelligent call semanteme communication method includes the following steps: that cloud server plays context voice by verbal system, and the call voice from verbal system is converted into call text by voice gateways, and the call text that will convert into is sent to the cloud server;The cloud server chooses primary keyword and corresponding keyword arranged side by side for matching corresponding default question letters according to the call text step by step from problem base: if successful match, the corresponding answer tone frequency of the default question letters of successful match is then sent to verbal system, is played out by verbal system;If it fails to match, preset audio of not understanding is sent to verbal system.The present invention also provides a kind of intelligent conversational systems based on intelligent call semantics recognition communication method.
Description
Technical field
It is particularly related to a kind of intelligent call semantics recognition communication method and intelligent conversational system.
Background technique
In order to reinforce the communication with client and prospect and exchange, larger production market propaganda dynamics provides more for client
Good product and technological service, many businessmans is all specially provided with customer service and the sales department of oneself, the work people of these departments
The workload that member links up service with customer phone daily is very big, carries out repetition, cumbersome problem answer, and cannot be daily
It all answers online within 24 hours or the consulting on duty for receiving user, trade company needs to pay huge human cost thus.Therefore, traditional
Single artificial customer service form can not adapt to the demand of user and the needs of enterprise's control cost.
With the continuous development of internet and the communication technology, people start by sight turn to human-computer intellectualization customer service and
Call form.But current human-computer interaction form is relatively simple, and user is typically likely encountered answering for problem by trade company
Case is preset in the server with written form, and when user inputs corresponding problem, server issues preset answer
User.Above-mentioned customer service form intelligence degree is low, and is that main answer-mode is not intuitive enough with text, and the scope of application is small,
People can be allowed to feel uninteresting over time.
Summary of the invention
It is an object of the invention in view of the drawbacks of the prior art or problem, provide a kind of intelligent call semantics recognition exchange
Method and intelligent conversational system.
Technical scheme is as follows: the problem of a kind of intelligent call semantics recognition communication method cloud server in library
Default question letters and answer audio corresponding with the default question letters are prestored, and are each default problem text
Word is equipped with multistage primary keyword and the arranged side by side keyword arranged side by side with primary keyword described in every level-one;The intelligent call semanteme exchange
Method includes the following steps: step 1: cloud server plays context voice by verbal system, and voice gateways will be from logical
The call voice of words equipment is converted into call text, and the call text that will convert into is sent to the cloud server;Step
2: the cloud server chooses primary keyword and corresponding key arranged side by side according to the call text step by step from problem base
Word is for matching corresponding default question letters: if successful match, the default question letters of successful match are corresponding
Answer tone frequency be sent to verbal system, played out by verbal system;If it fails to match, audio is not understood by preset
It is sent to verbal system.
Preferably, step 1 includes the following steps:
Step 1.1: cloud server plays context voice by verbal system, and receives the call language from verbal system
Sound;
Step 1.2: the call voice is converted into call text by voice gateways.
Preferably, answer library is additionally provided in the cloud server, prestored in the answer library default answer text and
Rhetorical question audio corresponding with the default answer text, step 2 include the following steps:
Step 2.1: cloud server chooses primary keyword and its keyword arranged side by side step by step from problem base, goes accurately to match
Call text, carries out step 2.2 if successful match, recycles if non-successful match and carries out step 2.1, if circulation terminates
Then carry out step 2.4;
Step 2.2: answer audio corresponding to the default question letters by successful match or rhetorical question audio are sent to network
Verbal system is played out by verbal system, if what is sent is rhetorical question audio, cloud server, which receives, comes from Internet phone-calling
The answer text of equipment, and carry out step 2.3;If what is sent is to answer audio, step 2.4 is carried out;
Step 2.3: cloud server chooses primary keyword and its keyword arranged side by side from the answer library of rhetorical question problem step by step,
It goes accurately to match and answers text;Return step 2.2 if successful match, if non-successful match, which recycles, carries out step 2.3, and
Step 2.4 is carried out if recycling and terminating;
Step 2.4: judging whether the answer audio sent is to terminate context, if not then choosing preset next section of context
Voice carries out step 2;If then cloud server terminates process, and passes through Internet phone-calling equipment kill-call.
Preferably, the context in the intelligent call semantics recognition communication method includes: prologue context, position in sequence
Context, price context invite context and terminate context.
A kind of intelligent conversational system using a kind of as above any intelligent call semantics recognition communication method includes:
Cloud server, verbal system and voice gateways;The cloud server plays call voice by the verbal system, described
The call text that call voice from the verbal system is converted by voice gateways, and the call text that will convert into is sent
To the cloud server moreover, the cloud server is chosen from problem base or answer library step by step according to the call text
Primary keyword and corresponding keyword arranged side by side are for matching corresponding default question letters: if successful match, general
It is sent to verbal system with the successful default corresponding answer audio of question letters or rhetorical question audio, is broadcast by verbal system
It puts;If it fails to match, preset audio of not understanding is sent to verbal system.
Technical solution provided by the invention has the following beneficial effects:
Intelligent call semantics recognition communication method proposed by the present invention and intelligent conversational system, enable intelligent call equipment
Intelligent " simulation virtual " customer service enough is provided for user, profession can not only be provided for user, and accurately call voice question and answer take
Business, and more understandable, humanized, vivid and interesting is replied compared to text.
Detailed description of the invention
Fig. 1 is that the embodiment of the present invention provides begin in a kind of intelligent call semantics recognition communication method context and position context
Schematic process flow diagram;
Fig. 2 is that the embodiment of the present invention provides a kind of process signal of price context in intelligent call semantics recognition communication method
Block diagram.
Fig. 3 is that the embodiment of the present invention provides the process signal that context is invited in a kind of intelligent call semantics recognition communication method
Block diagram.
Fig. 4 is that the embodiment of the present invention provides the process signal for terminating context in a kind of intelligent call semantics recognition communication method
Block diagram.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Although the step in the present invention is arranged with label, it is not used to limit the precedence of step, unless
Based on the execution of the order or certain step that specify step needs other steps, otherwise the relative rank of step is
It is adjustable.It is appreciated that term "and/or" used herein be related to and cover in associated listed item one
Person or one or more of any and all possible combinations.
It is prestored in library in intelligent call semantics recognition communication method provided by the invention, the problem of cloud server default
Question letters and answer audio corresponding with the default question letters, and multistage is equipped with for each default question letters
Primary keyword and the arranged side by side keyword arranged side by side with primary keyword described in every level-one.
Moreover, be additionally provided with answer library in the cloud server, prestored in the answer library default answer text and with
The corresponding rhetorical question audio of the default answer text, it should be understood that multistage is also equipped with to each default answer text
Primary keyword and the arranged side by side keyword arranged side by side with primary keyword described in every level-one.
It should be noted that the keyword arranged side by side is the similar import or equivalent in meaning and pronounce different of primary keyword
Word.
Specifically, the intelligent call semanteme communication method includes the following steps:
Step 1: cloud server plays context voice by verbal system, and voice gateways will be logical from verbal system
Language sound is converted into call text, and the call text that will convert into is sent to the cloud server;
Step 2: the cloud server chooses primary keyword and opposite according to the call text step by step from problem base
The keyword arranged side by side answered is for matching corresponding default question letters:
If successful match, the corresponding answer tone frequency of the default question letters of successful match is sent to call and is set
It is standby, it is played out by verbal system;If it fails to match, preset audio of not understanding is sent to verbal system.
In step 1, include the following steps:
Step 1.1: cloud server plays context voice by verbal system, and receives the call language from verbal system
Sound;
Step 1.2: the call voice is converted into call text by voice gateways.
In step 2, include the following steps:
Step 2.1: cloud server chooses primary keyword and its keyword arranged side by side step by step from problem base, goes accurately to match
Call text, carries out step 2.2 if successful match, recycles if non-successful match and carries out step 2.1, if circulation terminates
Then carry out step 2.4;
Step 2.2: answer audio corresponding to the default question letters by successful match or rhetorical question audio are sent to network
Verbal system is played out by verbal system, if what is sent is rhetorical question audio, cloud server, which receives, comes from Internet phone-calling
The answer text of equipment, and carry out step 2.3;If what is sent is to answer audio, step 2.4 is carried out;
Step 2.3: cloud server chooses primary keyword and its keyword arranged side by side from the answer library of rhetorical question problem step by step,
It goes accurately to match and answers text;Return step 2.2 if successful match, if non-successful match, which recycles, carries out step 2.3, and
Step 2.4 is carried out if recycling and terminating;
Step 2.4: judging whether the answer audio sent is to terminate context, if not then choosing preset next section of context
Voice carries out step 2;If then cloud server terminates process, and passes through Internet phone-calling equipment kill-call.
It should be noted that as shown in Figs 1-4, the context of the intelligent call semantics recognition communication method is wrapped in sequence
Include: prologue context, price context, invites context and terminates context at position context.
In the intelligent call semantics recognition communication method, after context of beginning exchanges successfully, position just can be successively carried out
Context, price context are set, context is invited and is exchanged with context is terminated.As long as prologue context, price context, is invited at position context
Context exchanges failed with terminate any context in context, then terminates whole flow process.
It may be selected to select ground, in the intelligent call semantics recognition communication method, default question letters and described pre-
If the primary keyword of answer text can be divided into second level, three-level or at least level Four, the present invention compare without limitation.
Moreover, being needed the primary keyword of every level-one and its corresponding keyword arranged side by side all in the retrieving of step 2
It is retrieved, only all keywords do not have matching result, can just obtain the result of non-successful match;As long as there is a key
Word successful match will then obtain the result of successful match.
A kind of intelligent conversational system using intelligent call semantics recognition communication method provided in an embodiment of the present invention includes:
Cloud server, verbal system and voice gateways;
The cloud server plays call voice by the verbal system, and institute's voice gateway will come from the call
The call text that the call voice of equipment is converted into, and the call text that will convert into is sent to the cloud server
Moreover, the cloud server chooses primary keyword from problem base or answer library according to the call text step by step
And corresponding keyword arranged side by side is for matching corresponding default question letters:
If successful match, the corresponding answer audio of the default question letters of successful match or rhetorical question audio are sent
To verbal system, played out by verbal system;If it fails to match, by it is preset do not understand audio and be sent to call set
It is standby.
A typical case place of the invention is to find more preferably substitution for some specific open source software to realize.Example
Such as: the open source application software of deep learning method base is depended on for some, it can be with frequency daily, weekly, monthly for this
A application software recommends new deep learning method base, if some the new deep learning method base recommended is than currently used
Deep learning method base has more methods or more preferably performance, then can be this new deep learning method base
It is imported into this application software as dependence, and removes or retain original deep learning method base.Based on theme correlation
Degree be open source software make recommendations, can preferably be realized in software development process code reuse, function add, realize it is excellent
The operation such as change, so that the development process for open source software provides help.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (5)
1. a kind of intelligent call semantics recognition communication method, it is characterised in that: prestored in library the problem of cloud server default
Question letters and answer audio corresponding with the default question letters, and multistage is equipped with for each default question letters
Primary keyword and the arranged side by side keyword arranged side by side with primary keyword described in every level-one;
The intelligent call semanteme communication method includes the following steps:
Step 1: cloud server plays context voice by verbal system, and voice gateways are by the call language from verbal system
Sound is converted into call text, and the call text that will convert into is sent to the cloud server;
Step 2: the cloud server chooses primary keyword and corresponding according to the call text step by step from problem base
Keyword arranged side by side is for matching corresponding default question letters:
If successful match, the corresponding answer tone frequency of the default question letters of successful match is sent to verbal system, by
Verbal system plays out;If it fails to match, preset audio of not understanding is sent to verbal system.
2. a kind of intelligent call semantics recognition communication method according to claim 1, which is characterized in that step 1 includes such as
Lower step:
Step 1.1: cloud server plays context voice by verbal system, and receives the call voice from verbal system;
Step 1.2: the call voice is converted into call text by voice gateways.
3. a kind of intelligent call semantics recognition communication method according to claim 1, which is characterized in that the cloud service
It is additionally provided with answer library in device, default answer text and corresponding with the default answer text anti-is prestored in the answer library
Ask that audio, step 2 include the following steps:
Step 2.1: cloud server chooses primary keyword and its keyword arranged side by side step by step from problem base, goes accurately to match call
Text carries out step 2.2 if successful match, recycled if non-successful match carry out step 2.1, if recycle terminate if into
Row step 2.4;
Step 2.2: answer audio corresponding to the default question letters by successful match or rhetorical question audio are sent to Internet phone-calling
Equipment is played out by verbal system, if what is sent is rhetorical question audio, cloud server, which receives, comes from Internet phone-calling equipment
Answer text, and carry out step 2.3;If what is sent is to answer audio, step 2.4 is carried out;
Step 2.3: cloud server chooses primary keyword and its keyword arranged side by side from the answer library of rhetorical question problem step by step, goes essence
Really text is answered in matching;Return step 2.2 if successful match, if non-successful match, which recycles, carries out step 2.3, and if
Circulation terminates then to carry out step 2.4;
Step 2.4: judging whether the answer audio sent is to terminate context, if not then choosing preset next section of context language
Sound carries out step 2;If then cloud server terminates process, and passes through Internet phone-calling equipment kill-call.
4. a kind of intelligent call semantics recognition communication method according to claim 2, which is characterized in that the intelligent call
Context in semantics recognition communication method includes: prologue context, position context, price context, invites context and knot in sequence
Beam context.
5. a kind of intelligent session system using a kind of intelligent call semantics recognition communication method as described in claim 1-4 is any
System, it is characterised in that: include: cloud server, verbal system and voice gateways;
The cloud server plays call voice by the verbal system, and institute's voice gateway will come from the verbal system
The call text that is converted into of call voice, and the call text that will convert into is sent to the cloud server
Moreover, the cloud server chooses primary keyword and phase from problem base or answer library according to the call text step by step
Corresponding keyword arranged side by side is for matching corresponding default question letters:
If successful match, the corresponding answer audio of the default question letters of successful match or rhetorical question audio are sent to logical
Equipment is talked about, is played out by verbal system;If it fails to match, preset audio of not understanding is sent to verbal system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810794548.6A CN108989589A (en) | 2018-07-19 | 2018-07-19 | A kind of intelligent call semantics recognition communication method and intelligent conversational system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810794548.6A CN108989589A (en) | 2018-07-19 | 2018-07-19 | A kind of intelligent call semantics recognition communication method and intelligent conversational system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108989589A true CN108989589A (en) | 2018-12-11 |
Family
ID=64550237
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810794548.6A Pending CN108989589A (en) | 2018-07-19 | 2018-07-19 | A kind of intelligent call semantics recognition communication method and intelligent conversational system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108989589A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110136703A (en) * | 2019-03-25 | 2019-08-16 | 视联动力信息技术股份有限公司 | A kind of fuzzy answer method and view networked system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103795876A (en) * | 2012-10-29 | 2014-05-14 | 殷程 | Intelligent voice script |
CN107566670A (en) * | 2017-09-05 | 2018-01-09 | 南京硅基智能科技有限公司 | The method that autonomous dialing, Intelligent dialogue and information analysis are realized based on speech recognition |
-
2018
- 2018-07-19 CN CN201810794548.6A patent/CN108989589A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103795876A (en) * | 2012-10-29 | 2014-05-14 | 殷程 | Intelligent voice script |
CN107566670A (en) * | 2017-09-05 | 2018-01-09 | 南京硅基智能科技有限公司 | The method that autonomous dialing, Intelligent dialogue and information analysis are realized based on speech recognition |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110136703A (en) * | 2019-03-25 | 2019-08-16 | 视联动力信息技术股份有限公司 | A kind of fuzzy answer method and view networked system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110390108B (en) | Task type interaction method and system based on deep reinforcement learning | |
JP7288446B2 (en) | Human-machine interaction method and electronic device | |
US10938988B1 (en) | System and method of sentiment modeling and application to determine optimized agent action | |
CA2929018C (en) | Natural expression processing method, processing and response method, device and system | |
CN110046230B (en) | Method for generating recommended speaking collection, and recommended speaking method and device | |
CN110413758B (en) | Session framework construction method and device based on machine learning | |
CN107741976B (en) | Intelligent response method, device, medium and electronic equipment | |
CN107889533A (en) | Management of dialogs data set provider | |
CN107818798A (en) | Customer service quality evaluating method, device, equipment and storage medium | |
CN110489198A (en) | A kind of method and system of worksheet | |
WO2017023566A1 (en) | Contact center virtual assistant | |
CN116235177A (en) | Systems and methods related to robotic authoring by mining intent from dialogue data using known intent of an associated sample utterance | |
CN116644145B (en) | Session data processing method, device, equipment and storage medium | |
US20190318742A1 (en) | Collaborative automatic speech recognition | |
CN116049360A (en) | Intelligent voice dialogue scene conversation intervention method and system based on client image | |
CN105279168A (en) | Data query method supporting natural language, open platform, and user terminal | |
CN109460503A (en) | Answer input method, device, storage medium and electronic equipment | |
CN117332072B (en) | Dialogue processing, voice abstract extraction and target dialogue model training method | |
CN109739969A (en) | Answer generation method and intelligent conversational system | |
CN115858744A (en) | Outbound method, device and storage medium based on AI | |
CN114548092A (en) | Customer service session scheduling method and device, equipment, medium and product thereof | |
CN108989589A (en) | A kind of intelligent call semantics recognition communication method and intelligent conversational system | |
CN112860873A (en) | Intelligent response method, device and storage medium | |
Surikova et al. | The Role of Artificial Intelligence in the Evolution of Brand Voice in Multimedia | |
CN111427444A (en) | Control method and device of intelligent device |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |