CN104580762B - A kind of method and apparatus that call forwarding is realized according to languages - Google Patents
A kind of method and apparatus that call forwarding is realized according to languages Download PDFInfo
- Publication number
- CN104580762B CN104580762B CN201310476446.7A CN201310476446A CN104580762B CN 104580762 B CN104580762 B CN 104580762B CN 201310476446 A CN201310476446 A CN 201310476446A CN 104580762 B CN104580762 B CN 104580762B
- Authority
- CN
- China
- Prior art keywords
- languages
- module
- caller
- call center
- recording file
- 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
Links
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a kind of method and apparatus that call forwarding is realized according to languages, the method comprising the steps of:Call Center Server is forwarded to the first agent phone by called, and communication process is recorded within a period of time;The first recording file is identified and judged using the first languages identification method;The first recording file is identified and judged using the second languages identification method;After call is completed in caller, prompt and receive the feedback information of the caller;The Calling Party Hooks On, Call Center Server record caller conversation number and corresponding agent phone.Implementing the beneficial effect of the inventive method and device is, similar languages service group can be automatically forwarded to automatically according to the category of language used in client, dual languages identification method is used simultaneously, while not only ensureing caller languages recognition accuracy, and simplify the identification step of conventional languages, save call center system resource.
Description
Technical field
The present invention relates to computer telephone integration field, and call forwarding is realized according to languages more specifically to one kind
Method and apparatus.
Background technology
Complained in order to more easily provide counseling services to passenger and effectively handle passenger, many enterprises are all logical at present
Cross and build enterprise call center to realize.
Modern call center, applying computer telephone integration (CTI) technology makes the service function of call center significantly
Strengthen.CTI technologies are using call voice as medium, and user can be by the button on telephone set come the calculating of operating call center
Machine.The mode at access calling center can be subscriber phone dial-up access, fax access, computer and modem
(MODEM) dial-up connection and Internet location (IP address) access etc., behind user's access calling center, with regard to calling can be received
Central task prompt tone, according to the voice message of call center, database just can be accessed, obtains required information service.And
The processing such as storage, forwarding, inquiry, exchange.
For call center according to languages forward call, now many entered using IVR (interactive voice response)
Row languages select, and such as put through the phone that a call center announces, it is necessary first to which button selects languages, and call center will be electricity
Words are transferred to corresponding languages service agent phone.But have it is many it is special connect the center of dealing with emergencies and dangerous situations, such as 110, do not allow to use IVR
Technology, but directly answer, if running into caller and called languages or dialect is different, the problem of can not linking up will be triggered.
However, in order to reach more preferable speech recognition effect, a kind of sample of language in speech recognition technology model library
Number will often reach 105More than, if it is desired to a variety of different languages of identification, then the requirement meeting to speech recognition technology model library
It is very high, and many samples are had due to being chronically at untapped state, cause the wasting of resources after renewal.
The content of the invention
The technical problem to be solved in the present invention is, for the drawbacks described above of prior art, there is provided a kind of real according to languages
The method and apparatus of existing call forwarding.
The technical solution adopted for the present invention to solve the technical problems is:Construct and a kind of realize call forwarding according to languages
Method, comprise the following steps:
S1:After caller is linked into Call Center Server, the Call Center Server is forwarded to the first seat by called
Phone, and communication process is recorded within a period of time, generate the first recording file;
S2:First recording file is judged using the first languages identification method;If being judged as the first languages,
Continue to allow the first seat telephone receiving, go to S4;If it is not, then go to S3;
S3:First recording file is identified using the second languages identification method, if being identified as in default languages
The second languages, then called will be forwarded to the second agent phone for answering the second languages, and go to S4;If the languages of user are not pre-
If one kind in languages, then recognition failures, go to step S5;
S4:After call is completed in caller, the Call Center Server is prompted and receives the feedback information of the caller, and root
According to feedback information amendment the first languages identification method or the second languages identification method;
S5:The Calling Party Hooks On, the Call Center Server record caller conversation number and corresponding agent phone, works as phase
When being linked into Call Center Server again with the caller of number, the agent phone of automatic transfer to last time.
In method of the present invention, in the step S1, communication process is carried out in first 20 seconds of air time
Recording.
In method of the present invention, in the step S2, recorded using the first languages identification method to described first
While file is judged, recording file is converted into by text message using ASR technology.
In method of the present invention, in the step S3, recorded using the second languages identification method to described first
After file identification failure, first recording file is uploaded to the webserver, the webserver returns to the first recording file
Identification information.
In method of the present invention, in the step S4, user's communication whole process is recorded, the recording of generation second
File, according to second recording file, the feedback opinion and the identification information, correct the first languages identification method
Or the second languages identification method.
Another aspect of the present invention, a kind of device that call forwarding is realized according to languages is constructed, for the language according to caller
Kind makes Call Center Server realize the function of automatic telephone switching, is arranged inside Call Center Server, the device includes
AM access module, interconnecting module, recording module, the first sound identification module, the second sound identification module, correcting module and storage mould
Block;
The AM access module is used to make caller be linked into Call Center Server;
The interconnecting module is used to make Call Center Server be forwarded to corresponding agent phone by called;
The recording module is used to record to communication process, and generates the first recording file;
First sound identification module is used to judge whether recording file is the first languages;
Second sound identification module is used to identify whether recording file is to preset one kind in languages;
The correcting module is used for according to feedback information amendment the first languages identification method or the second languages identification side
Formula;
The memory module be used for record caller conversation number and corresponding agent phone, when identical number caller again
It is secondary when being linked into Call Center Server, the agent phone of automatic transfer to last time.
In device of the present invention, the recording module is additionally operable to the recording life of the air time in 20 seconds before
The second recording file is generated into the first recording file, or according to the whole recording of call.
In device of the present invention, the device also includes being used to recording file is converted into text using ASR technology
The text conversion module of information.
In device of the present invention, the device also includes being used for the webserver and use for returning to the identification information
In the uploading module that first recording file is uploaded to the webserver.
Implement the method and apparatus that call forwarding is realized according to languages of the present invention, have the advantages that:Work as caller
When different with called languages, methods and apparatus of the present invention can be automatically forwarded to similar languages service according to the languages of client
Group, meanwhile, using dual languages identification method, while not only ensureing caller languages recognition accuracy, and simplify common-use words
The identification step of kind, saves call center system resource.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:
Fig. 1 is a kind of flow for method that call forwarding is realized according to languages that a preferred embodiment of the present invention provides
Figure;
Fig. 2 is a kind of square frame for device that call forwarding is realized according to languages that a preferred embodiment of the present invention provides
Figure;
Fig. 3 is a kind of flow for method that call forwarding is realized according to languages that another preferred embodiment of the invention provides
Figure;
Fig. 4 is a kind of square frame for device that call forwarding is realized according to languages that another preferred embodiment of the invention provides
Figure;
Fig. 5 is call center system schematic diagram.
Embodiment
In order to which technical characteristic, purpose and the effect of the present invention is more clearly understood, now compares accompanying drawing and describe in detail
The embodiment of the present invention.
As depicted in figs. 1 and 2, a kind of method that call forwarding is realized according to languages that one embodiment of the present of invention provides
And device, this method by Call Center Server according to the languages automatic transfer of caller to corresponding agent phone, the device can
To be arranged in Call Center Server, can also comprise the following steps separately as Call Center Server:
S1:After caller is linked into Call Center Server, Call Center Server is forwarded to the first agent phone by called,
And communication process is recorded within a period of time, generate the first recording file;Client is linked into enterprise by calling
Internal Call Center Server, Call Center Server are forwarded to the first agent phone, the first agent phone by called automatically
Operator carry out response, such as mandarin with local the most frequently used language, at the same time, Call Center Server also leads to user
Words recording, generates the first recording file, for identification.
S2:The first recording file is judged using the first languages identification method;If being judged as the first languages, continue
Allow the first seat telephone receiving, go to S4;If it is not, then go to S3;Speech recognition technology mainly includes feature extraction, pattern match
And three aspects of model library, the technology can be by voice input signal by feature extraction, into model library searching, then enters row mode
Matching, finally gives recognition result.If client will not be spoken standard Chinese pronunciation, then the operator of the first agent phone can not just be carried out
Link up or preferably link up, then now need to be forwarded to other agent phones, can be only achieved more preferable communication, now by the
One languages identification method None- identified recording file, then it is identified using another languages identification method.If described in client
It is exactly mandarin, then continue to allow the operator of the first agent phone to continue to answer.This step simplifies the knowledge of conventional languages
Other step, saves call center system resource.
S3:The first recording file is identified using the second languages identification method, if being identified as the in default languages
Two languages, then it will be called and be forwarded to the second agent phone for answering the second languages, go to S4;If the languages of user are not default languages
One kind in kind, then recognition failures, go to step S5;, can be by prestoring identification English such as what client said is English
Model library, the second languages identification method can know that this is English from default languages, then are forwarded to the second seat electricity by called
Words, the operator of the second agent phone can carry out English with client well and exchange.Languages as described in client are not advance
Set in the model library of speech recognition, then will be unable to identify.
S4:After call is completed in caller, Call Center Server is prompted and receives the feedback information of caller, and according to feedback letter
Breath the first languages identification method of amendment or the second languages identification method;Client be near completion call when, call center is hung up accordingly
Agent phone, and the feedback information of the system is improved according to client's languages from trend customer prompts button or phonetic entry, such as
One mother tongue is Japanese but call forwarding has been arrived seat electricity corresponding to English by the English-speaking client of meeting, Call Center Server
Words, then this client wish can according to the English-speaking accent of Japanese, next time connect phone when, if Japlish,
So it is automatically forwarded to agent phone corresponding to Japanese.Thus the first languages identification method and the second languages identification method are carried
Higher requirement is gone out, the first languages identification method can also be arranged to accent recognition mode.
S5:The Calling Party Hooks On, Call Center Server record caller conversation number and corresponding agent phone, works as same No.
When the caller of code is linked into Call Center Server again, the agent phone of automatic transfer to last time.Client wishes next energy
Associated traffic person can be connected without secondary call forwarding, then can be according to the access phone of client, or client is voluntarily
The phone of switching is set, and when dialing again, Call Center Server is with regard to automatic transfer to corresponding agent phone.If in calling
The languages of central server None- identified client, after client's on-hook, then according to recording file, pass through manual analysis or its other party
Formula, increase the method that can identify the languages.
On the other hand, a kind of device that call forwarding is realized according to languages is constructed, makes calling for the languages according to caller
Central server realizes the function of automatic telephone switching, is arranged inside Call Center Server, it is characterised in that the device bag
Include AM access module 1, interconnecting module 2, recording module 3, the first sound identification module 4, the second sound identification module 5, correcting module 6
With memory module 7;
AM access module 1 is used to make caller be linked into Call Center Server;
Interconnecting module 2 is used to make Call Center Server be forwarded to corresponding agent phone by called;
Recording module 3 is used to record to communication process, and generates the first recording file;
First sound identification module 4 is used to judge whether recording file is the first languages;
Second sound identification module 5 is used to identify whether recording file is to preset one kind in languages;
Correcting module 6 is used for according to feedback information amendment the first languages identification method or the second languages identification method;
Memory module 7 be used for record caller conversation number and corresponding agent phone, when identical number caller again
When being linked into Call Center Server, the agent phone of automatic transfer to last time.
Corresponding program is installed by interior portion in a call with upper module to can be achieved, can also set one it is special solely
Vertical equipment out, these functions are realized, by being connected with Call Center Server the corresponding interface, to realize the side of the present invention
Method.
Advantage of this embodiment is that (announce multiple telephone numbers without externally announcing multiple telephone numbers and be unfavorable for public affairs
Take charge of image display), without interactive voice selection languages are carried out, this method is by way of speech recognition, automatic telephone switching
To optimal agent phone.
As shown in figure 3, a kind of method that call forwarding is realized according to languages that an alternative embodiment of the invention provides, should
The method difference that the relatively upper embodiment of method provides is:Unidentified recording file can be uploaded to the webserver 10,
The webserver 10 returns to corresponding identification information, and the recording file of identification is converted into corresponding text message, is shown in
In first agent phone, allow the operator of the first agent phone according to the judgement of oneself, manual forward call, while manually
Correct the first languages identification method and the second languages identification method.
Installation drawing corresponding to this method is Fig. 4, and Fig. 4 is based on a upper embodiment, is added for that will be recorded using ASR technology
File translations are text conversion module 8, the uploading module 9 for the first recording file to be uploaded to network and the use of text message
In the webserver 10 for returning to identification information, the webserver 10 is not arranged in call center, but passes through remote rack
If the mode of the webserver 10, identification information is returned.
Fig. 5 is call center system schematic diagram, and in conjunction with Fig. 5, the method provided the present embodiment is explained, this method
Comprise the following steps:
S301:After caller is linked into Call Center Server by PSTN, Call Center Server called will be forwarded to the
One agent phone, and communication process is recorded in first 20 seconds of air time, generate the first recording file;Due to voice
Identification needs to carry out sampled voice, by feature extraction, then carries out characteristic matching according to model library and just can recognize that voice, institute
It is very helpful to the identification of whole languages to obtain recognition result in alap record length, and in the voice of 20 seconds,
The recognition accuracy of current speech identification technology can generally reach more than 60%.
S302:The first recording file is judged using the first languages identification method, while will be recorded using ASR technology
File translations are text message;If being judged as the first languages, continue to allow the first seat telephone receiving, go to S304;If it is not, then
Go to S303;Different according to the object of identification, voice recognition tasks can substantially be divided into 3 classes, i.e. isolated word recognition (Isolated
Word Recognition), keyword identification (or keyword spotting, Keyword Spotting) and continuous speech recognition.
Wherein, the task of isolated word recognition is to identify previously known isolated word, such as " start ", " shutdown ";Continuous speech recognition
Task be then the arbitrary continuous speech of identification, such as a sentence or one section of word;And the keyword identification pin in continuous speech stream
To be continuous speech, but it and nonrecognition whole word, and simply detect known some keywords and wherein occur such as exist
" computer ", " world " the two words are detected in one section of word.First languages identification method will use keyword to identify, that is, identify
Whether client's languages are languages that the first seat phone traffic person can link up.Recording file is converted into by ASR technology
Text message, in the case of necessary, text message can be included in the first agent phone, allow the traffic of the first agent phone
Member carries out call forwarding according to the judgement of oneself.
S303:The first recording file is identified using the second languages identification method, if being identified as in default languages
Second languages, then it will be called and be forwarded to the second agent phone for answering the second languages, go to S304;If the languages of user are not pre-
If one kind in languages, then recognition failures, are uploaded to the webserver 10 by the first recording file, go to step S305;Second
Languages find the languages of matching, such as by the way of continuous speech recognition in the model library of speech recognition in default languages
Fruit can not identify, then be uploaded to Internet (i.e. the webserver) by LAN (Local Area Network).Such as Fig. 5 institutes
Show, agent phone can increase as needed, such as:First seat is " mandarin ", and the second seat is " English ", and the 3rd seat is
" Japanese " etc..
S304:After call is completed in caller, prompt and receive the feedback information of caller, by the whole recording generation of user's communication the
Two recording files, according to the second recording file and feedback opinion, correct the first languages identification method or the second languages identification method;
Feedback opinion and recording file, also network-feedback all should be used as correcting the first languages identification method or the second languages identification method
Reference information.
S305:The Calling Party Hooks On, Call Center Server record caller conversation number and corresponding agent phone, when identical
When the caller of number is linked into Call Center Server again, the agent phone of automatic transfer to last time.Especially as " 110 " this
The urgent phone of sample, as far as possible reduce calling steps be only it is most important.If the time is wasted on existing IVR interactive voices,
Or on forward call, then loss is very big caused by wherein.The present invention is reduced client and handed over by automatic identification languages
Mutual link, resource utilization ratio is improved, reach the purpose of automatic telephone switching.
Embodiments of the invention are described above in conjunction with accompanying drawing, but the invention is not limited in above-mentioned specific
Embodiment, above-mentioned embodiment is only schematical, rather than restricted, one of ordinary skill in the art
Under the enlightenment of the present invention, in the case of present inventive concept and scope of the claimed protection is not departed from, it can also make a lot
Form, these are belonged within the protection of the present invention.
Claims (2)
1. a kind of method that call forwarding is realized according to languages, allow Call Center Server according to the languages automatic transfer of caller extremely
Corresponding agent phone, wherein, Call Center Server is connected with the first agent phone for answering the first languages and answers the
Second agent phone of two languages, it is characterised in that comprise the following steps:
S1:After caller is linked into Call Center Server, the Call Center Server is forwarded to the first agent phone by called,
And communication process is recorded within a period of time, generate the first recording file;Described a period of time is in the air time
First 20 seconds;
S2:First recording file is identified and judged using the first languages identification method;If being judged as the first languages,
Then continue to allow the first seat telephone receiving, go to S4;If it is not, then go to S3;Using the first languages identification method to described first
While recording file is judged, recording file is converted into by text message using ASR technology;
S3:First recording file is identified using the second languages identification method, if being identified as the in default languages
Two languages, then it will be called and be forwarded to the second agent phone for answering the second languages, go to S4;If the languages of user are not default languages
One kind in kind, then recognition failures, go to step S5;First recording file is identified using the second languages identification method and lost
After losing, first recording file is uploaded to the webserver, the webserver returns to the identification information of the first recording file;
S4:After call is completed in caller, the Call Center Server is prompted and receives the feedback information of the caller, and according to institute
State feedback information amendment the first languages identification method or the second languages identification method;User's communication whole process is recorded, generated
Second recording file, according to second recording file, the feedback opinion and described identification information, correct first language
Kind identification method or the second languages identification method;
S5:The Calling Party Hooks On, the Call Center Server record caller conversation number and corresponding agent phone, works as same No.
When the caller of code is linked into Call Center Server again, the agent phone of automatic transfer to last time.
A kind of 2. device that call forwarding is realized according to languages, for making Call Center Server realize certainly according to the languages of caller
The function that turn is answered the call, is arranged inside Call Center Server, it is characterised in that the device includes AM access module (1), turned
Connection module (2), recording module (3), the first sound identification module (4), the second sound identification module (5), correcting module (6) and deposit
Store up module (7);The device also includes being used for the text conversion module that recording file is converted into text message using ASR technology
(8);The device is also including for returning to the webserver (10) of the identification information and for the first recording file to be uploaded
To the uploading module (9) of the webserver (10);
The AM access module (1) is used to make caller be linked into Call Center Server;
The interconnecting module (2) is used to make Call Center Server be forwarded to corresponding agent phone by called;
The recording module (3) is used to record to communication process, and generates the first recording file;The recording module is also used
The first recording file is generated in the recording of the air time before in 20 seconds, or according to the whole record of recording generation second of call
Sound file;
First sound identification module (4) is used to judge whether recording file is the first languages;
Second sound identification module (5) is used to identify whether recording file is to preset one kind in languages;
The correcting module (6) is used for feedback information amendment the first languages identification method or the second languages identification side according to caller
Formula;
The memory module (7) be used for record caller conversation number and corresponding agent phone, when identical number caller again
It is secondary when being linked into Call Center Server, the agent phone of automatic transfer to last time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310476446.7A CN104580762B (en) | 2013-10-12 | 2013-10-12 | A kind of method and apparatus that call forwarding is realized according to languages |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310476446.7A CN104580762B (en) | 2013-10-12 | 2013-10-12 | A kind of method and apparatus that call forwarding is realized according to languages |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104580762A CN104580762A (en) | 2015-04-29 |
CN104580762B true CN104580762B (en) | 2018-01-05 |
Family
ID=53095888
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310476446.7A Active CN104580762B (en) | 2013-10-12 | 2013-10-12 | A kind of method and apparatus that call forwarding is realized according to languages |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104580762B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6356206B2 (en) * | 2016-12-15 | 2018-07-11 | 東芝エレベータ株式会社 | Elevator emergency call system |
CN107995376A (en) * | 2017-11-07 | 2018-05-04 | 平安科技(深圳)有限公司 | A kind of user reports a case to the security authorities processing method and terminal device |
CN110166950B (en) * | 2018-02-13 | 2022-06-24 | 海能达通信股份有限公司 | Answering feedback method, related equipment and system |
CN111277712A (en) * | 2019-07-11 | 2020-06-12 | 上海联逾信息技术有限公司 | Telephone calling-out system based on language type classification processing |
CN111277713A (en) * | 2019-07-11 | 2020-06-12 | 上海联逾信息技术有限公司 | Customer service switching system based on telephone call |
CN110675856A (en) * | 2019-10-14 | 2020-01-10 | 苏州思必驰信息科技有限公司 | Man-machine conversation method and device for call center |
CN112141845B (en) * | 2020-09-28 | 2022-08-02 | 上海三菱电梯有限公司 | Elevator emergency call system and method |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010002973A (en) * | 2008-06-18 | 2010-01-07 | Oki Electric Ind Co Ltd | Voice data subject estimation device, and call center using the same |
CN101478613B (en) * | 2009-02-03 | 2011-11-30 | 中国电信股份有限公司 | Multi-language voice recognition method and system based on soft queuing call center |
CN102802114B (en) * | 2012-06-20 | 2015-02-18 | 北京语言大学 | Method and system for screening seat by using voices |
-
2013
- 2013-10-12 CN CN201310476446.7A patent/CN104580762B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN104580762A (en) | 2015-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104580762B (en) | A kind of method and apparatus that call forwarding is realized according to languages | |
CN106202301B (en) | A kind of intelligent response system based on deep learning | |
US10171660B2 (en) | System and method for indexing automated telephone systems | |
US6456619B1 (en) | Method and system for supporting a decision tree with placeholder capability | |
US7933399B2 (en) | System and method for utilizing virtual agents in an interactive voice response application | |
US7460652B2 (en) | VoiceXML and rule engine based switchboard for interactive voice response (IVR) services | |
US6587558B2 (en) | System and method for virtual interactive response unit | |
US7751551B2 (en) | System and method for speech-enabled call routing | |
US8515028B2 (en) | System and method for externally mapping an Interactive Voice Response menu | |
US8625749B2 (en) | Content sensitive do-not-disturb (DND) option for a communication system | |
CN111885272A (en) | Intelligent call-out method for supporting telephone by call center seat and intelligent call center system | |
US20070135101A1 (en) | Enhanced visual IVR capabilities | |
US20040203660A1 (en) | Method of assisting a user placed on-hold | |
CN101478613B (en) | Multi-language voice recognition method and system based on soft queuing call center | |
MX2007002299A (en) | Method, system and software for implementing an automated call routing application in a speech enabled call center environment. | |
CN110392168B (en) | Call processing method, device, server, storage medium and system | |
WO2003088211A1 (en) | Dynamic and adaptive selection of vocabulary and acoustic models based on a call context for speech recognition | |
CN112866086A (en) | Information pushing method, device, equipment and storage medium for intelligent outbound | |
CN1863254A (en) | Method and system of call central service | |
US20080086690A1 (en) | Method and System for Hybrid Call Handling | |
JP2008118511A (en) | Telephone reception support device | |
JP2016225740A (en) | Speech communication distribution system, call control device and program | |
US20040109543A1 (en) | Method of accessing an information source | |
EP2469823B1 (en) | Computer telecommunication integration exchanger (ctiex), system and method for channel associated data transmission of agent and automatic service | |
JP2009290532A (en) | Call receiving system, call receiving method and call receiving program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |