CN101572756A - Dialogue system and voice dialogue processing method - Google Patents
Dialogue system and voice dialogue processing method Download PDFInfo
- Publication number
- CN101572756A CN101572756A CNA2008100955873A CN200810095587A CN101572756A CN 101572756 A CN101572756 A CN 101572756A CN A2008100955873 A CNA2008100955873 A CN A2008100955873A CN 200810095587 A CN200810095587 A CN 200810095587A CN 101572756 A CN101572756 A CN 101572756A
- Authority
- CN
- China
- Prior art keywords
- voice
- dialogue
- receiver
- speech
- signal
- 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
Images
Abstract
The invention relates to a dialogue system and a voice dialogue processing method. In the voice dialogue system, the method comprises the following steps: acquiring data of a specific person, then communicating a receiver through the data of the specific person, and generating a voice querying signal according to the data of the specific person; transferring the voice querying signal to the receiver who makes a reply according to the voice querying signal to generate a voice replying signal; acquiring a voice corresponding signal after the identifying process is carried out on the voice replying signal; carrying out the procedure for confirming the identity of the specific person according to the voice corresponding signal, and generating a dialogue controlling signal according to the result (pass/fail/uncertainty) of the procedure for confirming the identity of the specific person to generate a voice responding signal; and transferring the voice responding signal to the receiver to continue or end dialogue. Only after the receiver is confirmed to be the specific person, the information transmission procedure can be carried out, and then, the information to be notified to the specific person is transferred to the receiver.
Description
Technical field
The present invention relates to a kind of conversational system, and particularly relevant for a kind of person's that confirms to be talked about the object speech dialogue system and voice dialogue processing method.
Background technology
Gradually becoming the modern society of main flow in automation service, utilize voice system that option is provided, the client just can be finished by simple button several times want the item handled, has been very common a kind of automation method of service.Yet, because voice system can take many times when prompt options is more, even it is bad and cause unclear part to have reception at communication process, so for the client, just the too high problem of waits for too long and cost of the phone call is arranged, and for the system merchant that voice service is provided, then might cause other client to dial in, and then must increase telephone door number or face the awkward situation of being complained by the client because part customer use circuit cross for a long time.
With the funds on account collection is example, because content relates to privacy, the dealer sees through personnel with the Advise By Wire collection, perhaps the news in brief (SMS) of hypothesis mobile phone or voice mail has only me can learn under the prerequisite, transmission news in brief or voicemail messages earlier.But news in brief or the message can't confirm whether the client learns, and the degree of learning what.Moreover, still there is the people not see news in brief in the society and listens to voice mail, the real-time whether complete information conveyed of wanting of hearing of end client that grasps when its effect that can reach just can't be as Advise By Wire.Relative Advise By Wire, use news in brief or message to more not hommization of user, and the problem of telephone conversation notice with regard to there not being voice system can not use, also can obtain client's response at once, need not wait action that the client does answer especially as telegram in reply/letter in reply/return news in brief etc., can not take other method even have contact go up the client at once yet, obtain contact as people finder's mode and client.
Existing technology is user of service's notice, and its cost is higher on the one hand, and personnel have problem on and off duty, class time of having a rest of having a meal, having a holiday, and the service of full-time intersexuality can't be provided.On the other hand, though the employing personnel relatively have elasticity with respect to conversational system, the reaction of relatively being in a bad mood comparatively speaking, and different people service quality has and can't reach conforming problem.In addition, if the mode of employing personnel notice relates to privacy information owing to pass on possibly, jeopardize client's individual privacy relatively.
In addition, again with promotion business, for example provide the client about promotional information such as low-interest loan or the processing of circulation interest, the phone of possible cracking down illegal activities Chang Duotong has only several clients that take on the telephone interesting.The mode of employing personnel inquiry also is to cause quite high cost, and above-noted persons' problem.
So some work of comparing routine (Routine) or relating to privacy information notice on the trend, gradually all wish and to carry out preliminary operation by elder generation earlier by machine, and manpower is concentrated on the client that remaining machine can not be handled or machine processing is bad as far as possible.Because news in brief or message can't be grasped the whether complete information conveyed of wanting of receiving of client really, the effect comparison can't be grasped.
Existing conversational system is used for being subjected to words side (Inbound) more, if be used for the outer system (Outbound) of dialling, also mostly is that the information reception and registration or the suggestion of special object are not collected in election advertisement or market survey etc. with for example using.Because whether do not have to confirm is specific people's mechanism,, can not be used for transmitting the information that relates to privacy so can't confirm whether to have the problem of sample repetition.More need not propose the dialogue with before, be used for tracking and grasp for subsequent job.
Even owing to used personnel correctly to pass on information, still might not achieve the goal (still not going to pay) as client's debt, talk with conversational system still more, also may run into can't identification, or client refusal and machine talk with and problem such as directly hang up, and stay the client that some need the tracking and the grasp of subsequent job.
Many patents propose the similarly outer system of dialling, for example broadcast system concentrates on outer (Outbound) phone of dialling of connection and gives the method that is subjected to words side (Inbound) in addition, as the U.S. the 7th, the technology that patent proposed of 230, No. 927 " Non-blocking expandable call center architecture " promptly belongs to this type of.In addition, the patent as the U.S. the 7th, 215, No. 744 " Call processing systemwith call screening " proposes the method that the out-going call list is wanted in a kind of screening.
In addition, for example in the U.S. the 6th, 574, the patent of No. 599 " Voice-recognition-basedmethods for establishing outbound communication through a unifiedmessaging system including intelligent calendar interface " proposes the task assignment system of the Orion system of the similar MIT of the Massachusetts Institute of Technology, or in the U.S. the 6th, 885, the patent proposition of No. 734 " System and method for the creation and automaticdeployment of personalized; dynamic and interactive inbound andoutbound voice services, with real-time interactive voice databasequery " is about replying the method that is subjected to words side (Inbound) inquiry by outer dial (Outbound).
In addition, about the best (Best Time To Call that makes a phone call the time, the beneath BTTC that is called for short) technology, for example can be with reference to the U.S. the 5th, the patent of 742, No. 674 " Automatic call-back system andmethod using data indicating best time to call ", or in the U.S. the 6th, the patent of 917, No. 671 " Network-based method for notification of the besttime to call ".Above-mentioned traditional BTTC technology all is to allow system the record earlier of the demand that will dial, by the time help to dial that words side dials and clawback connects and dials words side and finish conversation the most cheap period again, but do not have when the record that contacted and finished at the past is set or which client of that telephone number making call is the function of relatively more suitable time.
Summary of the invention
The invention provides a kind of conversational system and voice dialogue processing method, can confirm automatically whether present those who answer is the specific people that we will look for.
The invention provides a kind of conversational system and voice dialogue processing method, all that dial specific people outward can be answered situation, each use the behavior coded representation, to calculate the best dial-out time of the outer specific people of dialling next time.
Based on above-mentioned purpose, the present invention proposes a kind of speech dialogue system, comprises dialogue control module, voice identification module and voice generation module.This dialogue control module is in order to obtaining a specific personal data, and according to this specific personal data after being communicated with, obtain a receiver dialogic voice data.And voice identification module is connected to dialogue control module, in order to receive this dialogic voice data, produces a voice respective signal after identification.And this dialogue control module produces a dialogue control signal according to the voice respective signal.This voice generation module is connected to dialogue control module, produce an inquiry voice signal corresponding to this dialogue control signal, and dialogue control module sends the receiver to thus, is replied according to the inquiry voice signal by the receiver.
Based on above-mentioned purpose, the present invention proposes a kind of speech dialogue system, comprises dialogue manager, voice identification module and voice generation module.This dialogue manager produces the inquiry voice signal according to specific personal data in order to obtaining specific personal data, and according to specific personal data with after a receiver is communicated with, this is inquired that voice signal sends this receiver to; Wherein, reply the back by the receiver according to this inquiry voice signal and produce a reply voice signal, after carrying out identification at the reply voice signal, obtain a voice respective signal, carry out specific people's identity according to the voice respective signal and confirm flow process, and according to specific people's identity confirm flow process the result (I/non-I/uncertain) produce a dialogue control signal, and produce a voice responsive signal according to this, send the receiver to and continue identity validation or end dialog.This specific people's identity is confirmed flow process only after confirming that the receiver is exactly specific people, just meet specific people's identity after, just carry out the information transmission procedure, will send the receiver to corresponding to specific people's information.
Based on above-mentioned purpose, the present invention proposes a kind of speech dialogue system, comprises dialogue control module and voice generation module.This dialogue control module is in order to obtaining a specific personal data, and is communicated with a receiver according to specific personal data.The voice generation module is connected to dialogue control module, produces an inquiry voice signal corresponding to specific personal data, and sends the receiver to by dialogue control module, and obtains the voice response signal that the receiver responds the inquiry voice signal.This dialogue control module is carried out specific people's identity affirmation flow process at receiver's voice response signal, after confirming that the receiver is exactly specific people, carries out the information transmission procedure, will send the receiver to corresponding to specific people's information.
The present invention proposes a kind of voice dialogue processing method, comprise obtaining specific personal data, and according to specific personal data via after being communicated with, and according to specific personal data producing an inquiry voice signal, and send the receiver to; Wherein, reply the back by the receiver according to this inquiry voice signal and produce a reply voice signal, after carrying out identification at this reply voice signal, obtain a voice respective signal, carry out specific people's identity according to this voice respective signal and confirm flow process, and according to specific people's identity confirm flow process the result (I/non-I/uncertain) produce a dialogue control signal, and produce a voice responsive signal according to this, and continue specific people's identity and confirm or end dialog by sending the receiver to.
Conversational system proposed by the invention and voice dialogue processing method, at the mode of confirming specific people, except adopting the vocal print feature mode, can adopt the adding further feature to compare, such as obtain described specific people's related data according to database, as the biological characteristic at sex or age or the like or as described in specific people's habitual language or residence accent or the like, the data of utilizing above-mentioned feature and receiver to be answered are compared, to confirm whether to meet specific people's identity.
Outer conversational system and the voice dialogue processing method of dialling specific people proposed by the invention, can be in conjunction with graphical user's interface (Graphical User Interface), or optionally use recording system, allow the personnel that carry out subsequent job, for example negotiator waits and carries out ensuing business.
Because the record of dialog procedure, the outer conversational system of dialling specific people proposed by the invention can make the content that the personnel selection client of subsequent job replys and analyzes, and can save artificial expense in a large number.In addition,, can when talking with, specific people look for the conversation content in his past to adjust speech model and language model etc., so that improve the discrimination power of voice according to these records.
In addition, because the record of dialog procedure is found to differ very big with track record if the personnel of subsequent job compare the vocal print of client's answer, warn the situation generation that the related personnel has personation this moment.
The outer conversational system of dialling specific people proposed by the invention, more added make a phone call the time technology of (Best Time To Call) of so-called the best, through record after a while, can calculate the optimal contact of the listener time by statistical observation, the success rate of dialing after making can significantly promote.
Outer conversational system and the voice dialogue processing method of dialling specific people proposed by the invention, when the information of carrying out transmission procedure, in the time of will sending the receiver to corresponding to specific people's information, further detect unusual condition, when situation takes place if note abnormalities, then reenter specific people's identity and confirm flow process.
Above-mentioned unusual condition for the vocal print of finding this receiver not simultaneously, assert that then this unusual condition takes place.Or the answer of finding this receiver continues set number of times can't identification the time, or can assert all that when a period of time does not have any response this unusual condition takes place.
Above-mentioned reenter specific people's identity and confirm flow process after, produce and the unduplicated inquiry voice signal inquiry of last inquiry voice signal receiver according to specific personal data.If behind this inquiry voice signal inquiry receiver, remain unfulfilled affirmation, then carrying out other unduplicated inquiry voice signal, till finishing affirmation.Excessively do not bothering under client's the consideration, still can not judge whether to be this man-hour in the inquiry again of presetting certain number of times, system can hang up after the listener apologizes.
For above-mentioned feature and advantage of the present invention can be become apparent, embodiment cited below particularly, and cooperate appended graphicly, be described in detail below.
Description of drawings
Fig. 1 is the system block diagrams that illustrates an embodiment of speech dialogue system of the present invention;
Fig. 2 is the system job schematic flow sheet that illustrates the speech dialogue system of the embodiment of the invention;
Fig. 2 A illustrates in the system job flow process of speech dialogue system of the embodiment of the invention, adds the schematic diagram of vocal print comparison step.
[primary clustering symbol description]
100: speech dialogue system
110: conversational system
120: database
130: graphical user's interface
140: dial and connect module
150: recording module
160: service network
Embodiment
Proposed by the inventionly cooperate database schema to go out a conversational system and voice dialogue processing method with conversational system.Whether the simulation personnel are specific people with the prompting sentence inquiry of phonetic synthesis, confirm whether to meet specific people's identity by speech recognition with the mode or the mode of indeterminate (Implicit) of clear and definite (Explicit) again.
When above-mentioned listener answered in the mode of clear and definite (Explicit), for example clearly replying him was people's (as answering: I am) that we will look for, and the vocal print feature is consistent with the record of passing by, and answered with regard to visual herewith specific people.And if the listener replys in the mode of indeterminate (Implicit), for example inquiry is subjected to whether specific for this reason man-hour (as: may I ask you is Mr. Wang Daming) of words side, mode that do not deny is answered (as:) to the client to compare neutral, and meets with regard to visual herewith specific people according to the vocal print feature again.In addition, when if the mode that the client does not deny to compare neutrality is answered, do not confirm to be subjected to whether specific for this reason people (as: may I ask you is Mr. Wang Daming) of words side though system puts question to, if language person vocal print is consistent with the record of passing by, can regard as is that this specific people answers.And system is that the information that relates to privacy just can be passed on by system under this specific people situation of answering the call in identification only.
The outer conversational system of dialling specific people proposed by the invention, at the mode of confirming specific people, except adopting the vocal print feature mode, can adopt the adding further feature to compare, and obtain specific people's related data according to database, as biological characteristic or specific people's the habitual language or the residence accent or the like at sex or age or the like, the data of utilizing above-mentioned feature and receiver to be answered are compared, to confirm whether to meet specific people's identity.
The outer conversational system of dialling specific people proposed by the invention, can be in conjunction with graphical user's interface (Graphical User Interface, beneath abbreviation GUI), or optionally use recording system, allow the personnel that carry out subsequent job, for example negotiator waits and carries out ensuing business.
Because the record of dialog procedure, the outer conversational system of dialling specific people proposed by the invention, the content that client's contact in the past that the personnel of subsequent job only need be listened needs defensive position to continue contact is replied and analyzing can be saved artificial expense in a large number.In addition,, can adjust speech model, so that improve the discrimination power of voice to the dialogue of finishing according to these records.
In addition, because the record of dialog procedure is found to differ very big with track record if the personnel of subsequent job compare the vocal print of client's answer, warn the situation generation that the related personnel has personation this moment.
The outer conversational system of dialling specific people proposed by the invention, more added make a phone call the time technology of (Best Time To Call, beneath abbreviation BTTC) of so-called the best, through record after a while, observe the optimal contact time of listener, make the success rate of dialing significantly to promote.For example some client may fix each of meeting Tuesday, so Tuesday, contact all can't be successful.So according to the record of dialing in the past, Friday morning can be found the people, therefore, these records just can be done the reference of BestTime To Call (BTTC), and the rate of dialing after making that reaches of touching improves.
According to foregoing, the present invention proposes a kind of speech dialogue system, comprises dialogue manager, voice identification module and voice generation module.This dialogue manager is in order to obtaining specific personal data, and according to specific personal data with after a receiver is communicated with, and according to specific personal data producing an inquiry voice signal, and send the receiver to; Wherein, reply the back by the receiver according to this inquiry voice signal and produce a reply voice signal, carry out identification at this reply voice signal after, obtain a voice respective signal, this voice respective signal can be pre-record voice or synthetic speech one of them.Carry out specific people's identity according to this voice respective signal and confirm flow process, and according to specific people's identity confirm flow process the result (I/non-I/uncertain) produce a dialogue control signal, and produce a voice responsive signal according to this, and continue specific people's identity affirmation flow process or end dialog by sending the receiver to., after confirming that the receiver is exactly specific people, carry out the information transmission procedure, will send the receiver to corresponding to specific people's information.The beneath conversational system that the outer specific people of dialling proposed by the invention will be described with the embodiment of reality.
Please refer to Fig. 1, is the system's block schematic diagram that illustrates the embodiment of conversational system of the present invention.This speech dialogue system 100 comprises conversational system 110, database 120, graphical user's interface (GUI) 130, dials and connects module 140, recording module 150, reaches service network 160.As shown in the figure, graphical user's interface (GUI) 130, recording module 150 are indicated by dotted line, mainly are because this this module is function optionally, illustrate in the lump in this adding, and so be not as limit.
This conversational system 110 comprises dialogue manager 111, voice identification module 112 and voice generation module 113.Above-mentioned voice identification module 112 for example is automatic speech recognizing (AutomaticSpeech Recognition, " ASR " as shown in the figure), and voice generation module 113 for example is text-to-speech (Text to Speech, " TTS " as shown in the figure) module, this is only in order to illustrate present embodiment, another kind of mode can be to choose corresponding voice from many voice of pre-recording, or synthetic needed voice, not as limit.And above-mentioned service network 160, for example be a kind of integrated service digital network (Integrated Service Digital Network, " ISDN " as shown in the figure) or a kind of Public Switched Telephone Network (Public Switched TelephoneNetwork, " PSTN " as shown in the figure).
In conversational system 110, dialogue manager 111, voice identification module 112 and voice generation module 113 are connected to each other.And database 120 is connected to each other with dialogue manager 111, graphical user's interface 130 and recording module 150.130 graphical user's interfaces and recording module 150 interconnect.Dialing and connecting module 140 then is to interconnect with dialogue manager 111, and is connected to service network 160 by recording module 150.150 of recording modules are to interconnect with dialogue manager 111.
In this speech dialogue system 100, provide the client's who needs contact name to reach and its master data by database 120, as phone, sex, age or the like.Conversational system 110 is via dialing and connecting or the client of mode such as network connection and appointment constitutes and leads to.The mode that this formation is led to is to see through to dial and connect module 140 in this embodiment, but is not as limit.This dials and connects module 140 and sees through service networks such as ISDN or the PSTN 160 outer clients that set aside appointment, and 150 of recording modules just need be recorded when the recording demand is arranged.Usually this recording module 150 is between dialing and connecting between module 140 and the service network 160, if service network 160 is the ISDN system, then can analyze online ISDN package is also recorded automatically, if service network 160 is the words of PSTN system, may be just with and the mode that connects drive by dialing and connecting module 140, the online voice of recording are stored outward group the time.
The dialogue (speech data) that conversational system 110 is come identification to understand to be received by voice identification module 112 is basically putd question to by voice generation module 113 synthetic prompting sentences and is guided the client to answer us and wants the problem of asking.The control of 111 responsible whole dialogue flow processs of dialogue manager.When the various differences of listener are answered, all to do the reply of a corresponding system of institute.
Dialing and connecting module 140 can add phone flow process analysis software (Call Progressanalysis, be called for short CPA) by phonecard and formed.When being responsible for outer dialling to the state of the side talked about; for example put through, hang up, spacing, engaged line, the words of stopping, facsimile machine or entering voice mail, recorder and Auto Attendant status detection such as replys; judge whether to be valid telephone, if allow system can filter out the situation that to put through specific people smoothly.
Just, then can directly cut off and dial and connect if receiver's state is ring unanswered or invalid telephone state.And in one embodiment, invalid telephone state comprises spacing, stop words, suspend user mode one of them.And ring unanswered's telephone state comprises in shutdown, unmanned response, the engaged line or in the default stand-by period ring unanswered one of them.
In addition, be communicated with successfully, then judge whether to enter voice mail again or recorder is left word if dial and connect.If detecting and being subjected to words side is automatic speech exchange or automatic keying exchange, then imports the voice name or key in extension and reenter dial-up program and phone process analysis and rejudge and dial and connect whether success of connection.To be subjected to words side be facsimile machine if detect this, because whether have only me just can learn the success of faxing, therefore can only hang up after the annotation of write-back storehouse.
Graphical 130 user's interfaces (GUI) be the phone record of dialing before afterwards having access to for the down-stream treatment people and taped conversations content by database 120 and recording module 150 and can be on demand output every statistical analysis report of dialing the result automatically.
Detail operations is as follows, and at first, dialogue manager 111 can utilize database 120 to obtain this outer specific people's of dialling personal data, then sees through and dials and connects module 140 dial-outs, with connectivity serving network 160.When the other side answered, dialogue manager 111 engaged in the dialogue with those who answer via voice identification module 112 and voice generation module 113.The work that while recording module 150 is also recorded.Whether dialogue manager 111 can inquiry those who answer be the outer specific people of dialling, when dialogue manager 111 can't determine whether the other side is to dial specific man-hour outside this, then heavily asks with different words arts, still can't determine then to hang up as heavily asking again and again after heavily asking certain number of times.Confirm that when dialogue control 111 those who answer is this specific man-hour, then transmission information is given this specific people.Can utilize native system to reach the effect the same by this with personnel's Advise By Wire.In other words, dialogue manager 111 can confirm automatically whether present those who answer is this specific people, as confirming then inquiry once more.When repeatedly inquiry still can't be confirmed, then after heavily asking certain number of times, cut off online.When confirming that the other side is specific man-hour, then transmission information is given the other side.
Fig. 2 is the system job schematic flow sheet that illustrates the speech dialogue system of the embodiment of the invention.At first, please in the lump with reference to Fig. 1 explanation, after system start-up, dialogue manager 111 can keep waiting for the state of database 120 inputs, up to as step S201, database 120 provides the client's who needs contact master data (ID Entry), as name, phone, sex or the like.Then as step S202, receive these master datas after, enter step S203 driving dial module 140 and dial.Dial and connect module 140 and can judge whether outer dialling is valid telephone by phone flow process analysis software (CPA).If be judged as invalid phone as step S207; for example be spacing, stop words, suspend use or the like state; or judgement learns to be states such as engaged line, ring unanswered, nothing response or shutdown as step S206; just can directly hang up; and the corresponding behavior code of annotation (Action Code) by this behavior code of step S212 write-back to database 120, and finish this and take on the telephone.
In conversational system proposed by the invention 100, the result that the flow process of each operation produces, can set corresponding specific behavior code (Action Code), and these behavior codes will be stored in database 120, and record as the whole dialogue system, so that can produce different application and design according to different behavior codes by system operator, this no longer superfluous stating.
Except step S207 is judged as invalid phone, or step S206 judges that learn can't communications and liaison, if connected but entered telecommunications service when dealer's voicemail service is provided as step S205, this moment, conversational system 100 can directly be written back to database 120 via step S212 without the message operation, and finished this and take on the telephone.Another mode then is to carry out message notification or to entertain guests the family telegram in reply through step S2051.Because client's voice mail may be set and must import password and could leave a message, or mailbox completely can't be left a message, when if conversational system 100 produces as the rumor unusual condition of step S2052 when finding message, can be when annotation behavior code (Action Code), according to the corresponding behavior code of message success or not difference annotation, and be written back to database 120, and finish this and take on the telephone by step S212.
Judge when having the people to answer if dial and connect module 140, then carry out specific people's identity affirmation flow process of step S2041 as step S204.Whether the simulation personnel can be specific people with the prompting sentence inquiry of phonetic synthesis, confirm whether be specific people by speech recognition with the mode or the mode of indeterminate (Implicit) of clear and definite (Explicit) again.And in this step, dial and connect module 140 and judge that dialing and connecting circuit has constituted communications and liaison and get final product, even but the someone has answered but and has not answered, whether also can enter inquiry is specific people's flow process, for example system determines to have obtained communications and liaison, but also silent after waiting for three seconds, then will directly enter the program of inquiry.
So-called mode with clear and definite (Explicit) answer be exactly the clear and definite expression of the people that receives calls he be the people that we will look for, for example:
Is system: may I ask you Mr. Wang Daming?
The client: I am
Mode that do not deny is answered (as:) if the client is to compare neutral, because if the information of notice is when having the consideration of privacy, this moment, 100 of conversational systems can be carried out identity validation with the mode of indeterminate (Implicit), just can confirm to be subjected to whether words side is specific people with clear and definite enquirement once more, for example further inquiry in conversational system 100:
Is system: may I ask you Mr. Wang Daming?
The client:
If whether the client is under the specific people clearly puing question to, still with relatively more neutral and mode that do not deny is answered, and sound characteristic meets, if just language person vocal print and the record in past are consistent.If or the database past do not have client's vocal print data, conversational system 100 can be utilized some specific biological characteristic, for example the sound pitch is judged whether sex meets etc. and is compared.
And conversational system 100 can be set the authentication grade that different indeterminate (Implicit) mode that needs is carried out identity validation according to the several applications needs.If meeting indeterminate (Implicit) authentication grade that sets still can treat as and meet this specific people's identity.If the result that specific people's identity affirmation program of client obtains reaches the authentication grade that conversational system 100 sets, be under this specific people situation of answering the call only just in identification, step S208, S209 under system just can proceed to and S210 pass on the information that relates to privacy, otherwise equally by step S212 write-back behavior code to database 120, and finish this and take on the telephone.
If specific people's identity of step S2041 confirms in the flow process that if the client responds with more uncertain answer, this moment, different words arts were constantly adopted in conversational system 100 meetings, guided the user to compare clear and definite answer again, rather than required the client to say again.On the one hand be that the content that conversational system can identification must be limited, with different talk about arts can avoid allowing the client always repeat same answer system can't identification.On the other hand, the client hears system one repetition of speaking out, and may take for system's entanglement, may directly not hang with regard to having patience to continue to say.
Because the client may hang at any time, if when detecting the client and hang up as step S2042, conversational system 100 can judge how much information the client hears as the time that step S2044 is hung by the client, the record that has just added the voice halts, for example conversational system 100 inquiries: be XX bank here, is may I ask you Mr. Wang Daming? if the client only hears XX bank, three words of Wang Daming are not all listened and have just been hung, basically all do not hear just to hang with client XX bank and may just represent different meanings, obviously the former client may be in the bank of hiding.At this moment step S2044 just can finish a little according to the meaning of one's words that system need set each enquirement sentence, sets different behavior codes.
If through step S2041, client identity confirm to finish by, that is to say specific people's authentication grade that the client sets by system, system just can be optionally many query users problem or directly transmit information again, if any the more problem of inquiry, then whether the affirmation client as step S2091 hangs up, or step S2091 and S2092, just as step 2042 or 2044, the capital checks whether the client hangs up, if and hang up, the conversational system 100 that whether tin is over before hanging is wanted the problem of puing question to, and has just added the detection and the record of voice halts.
At last, pass on if conversational system 100 is finished information, equally by step S212 write-back behavior code to database 120, and finish this phone of taking on the telephone.
Because even personnel have correctly passed on information, still might not achieve the goal (still not going to pay) as client's debt, talk with conversational system still more, also may run into and can't identification understand, or client's refusal and machine are talked with and directly hung problem such as electricity and stay the client that some need the subsequent job personnel to continue to follow the trail of.So this conversational system 100 also proposes to supply defensive position such as negotiator to do the method for tracking in conjunction with graphical user's interface (GUI) 130 and recording system 150.
The conversational system 100 of the embodiment of the invention, the obtained customer voice of voice identification module 112 is proposed, when the subsequent job personnel are needed, can only listen the part that the client speaks and filter the speech data of dialing procedure, or conversational system 100 puts question to the speech data of processes to replace or the like with literal.So, a general rule can just be heard out about 10 seconds as about 60 seconds phone.
The conversational system 100 of the embodiment of the invention, owing to stored the dialogue record at database 120, if therefore client's dialogic voice of once finishing can make the identification of next time more accurate for adjusting speech model.Or set specific identification program at the client, make specific people's identity of indeterminate (Implicit) confirm that flow process can finish smoothly faster.In addition also can be according to the record of dialing in the past, utilization used to contact this specific people's time, reference as Best Time To Call (BTTC), the rate of dialing after making that reaches of touching improves, even when the down-stream operating personnel carried out further tracking, system can be putting through back comparison client's vocal print automatically, if this client's sound and record before are far short of what is expected, just can remind personnel, the other side has the possibility of personation.
Whether the foregoing description can confirm specific for this reason people with the vocal print of collecting in the past is auxiliary except step S2041, even arrived step S208 and step S209, still can whether the situation of substituting and answering be arranged midway with the vocal print inspection, such as, the client that a past was not dialed, in the identity validation stage,, just can only judge whether specific for this reason people with the result of voice identification owing to there is not the vocal print record.It is other people's pickup phone that a kind of situation is arranged, behind the name that the system of hearing will look for, be given to another person and listen, and also may be another kith and kin of client and this people may be the client.And background voice in the middle of the commentaries on classics phone or noise may make system's erroneous judgement answer for me, so, shown in Fig. 2 A, if system has added vocal print comparison step S2081 between step S208 and S209, when finding vocal print and response before obviously not simultaneously, will stop transmission information, if the information of transmitting also is less than meaning of one's words starting point, as " may I ask you pays in? " pay in, system will get back to step S2041 and reaffirm identity, carry out behind the identity validation just transmission information equally again, otherwise, will directly enter step S212 annotation and may transmission information give non-this specific people, terminating telephone, or the switching true man, do follow-up processing by personnel.
Above-mentioned vocal print comparison step S2081, also can detect step by unusual condition replaces, this unusual condition detection mode, except as above-mentioned test returns vocal print unusually, but also detection of biological feature unusual and stop transmission information, and get back to step S2041 and reaffirm identity, for example be originally that male voice is changed to female voice.This unusual condition also the client for example repeat to reply fuzzy statements such as " " again and again with relatively more neutral and mode that do not deny repeats to answer again and again.Except above-mentioned example, system also can design different unusual conditions as required.
In addition, if when carrying out identity validation again, produce unduplicated other question sentence inquiry receiver according to this specific personal data.Just adopt different words arts heavily to ask, think that to avoid the side of word system works as machine and repeats identical dialogue again and again.
The outer conversational system of dialling specific people proposed by the invention, can be owing to be subjected to talk about because environment is noisy, or driving etc. and wholwe-hearted so that radio reception is more unclear, existing method of past all is that the client that asks of very intuition says once again, say once again but the client is more out of patience usually, may also think it has been system's entanglement, the family of how always calling out-customer is said once again.So can use different words arts instead heavily asks, rather than ask the client to say once again, main benefit is except allowing the client more can not think system's entanglement, also relatively can induce the client to answer with different numbers, identification system can have certain scope by identification grammer glossary after all, so can avoid allowing the client always repeat same answer system can't the identification number, and the identification success rate of dialing can significantly be promoted, and the ratio of being hung up by the client also can effectively reduce.Another may be this client at ordinary times just by his relatives on behalf of paying in, such as there is this client of record in system all is that Mr. is on behalf of processing, when detecting boy student's response, may just change ask may I ask " you are Mr. of Miss XXX? ", later on just can be with his Mr.'s vocal print inspection boy student's response.
In sum, the present invention can dial outward and confirm whether its those who answer is this specific people, and transmission information, to guarantee the secret of its information.And it is for utilizing systems response, not free restriction and the inconsistent problem of quality, can full-time intersexuality service is provided.And utilize the recording of behavior code and system can learn the situation of transfering to each time,, help personnel to understand its phychology, to improve the continuity of personnel's subsequent treatment in order to analyze the behavior pattern that it is answered.And the best that can learn this specific people is transferred to the time and answered custom, in order to avoid transfer to Nobody answers or engaged line.Moreover the investigation of the phone of routine can be replaced personnel with native system, also can bring into play its effect, but and on the screening phone list to the interest of certain product, improve the quality of product and sell rate.
Though the present invention discloses as above with embodiment; right its is not in order to limit the present invention; have in the technical field under any and know the knowledgeable usually; without departing from the spirit and scope of the present invention; when can doing a little change and retouching, so protection scope of the present invention is as the criterion when looking accompanying the claim person of defining.
Claims (66)
1. speech dialogue system is characterized in that comprising:
One dialogue control module, in order to obtaining a specific personal data, and according to this specific personal data after being communicated with, obtain a receiver dialogic voice data;
One voice identification module is connected to this dialogue control module, in order to receive this dialogic voice data, produces a voice respective signal after identification, and this dialogue control module produces a dialogue control signal according to this voice respective signal; And
One voice generation module, be connected to this dialogue control module, this voice generation module produces an inquiry voice signal corresponding to this dialogue control signal, and sends this receiver to by this dialogue control module, is replied according to this inquiry voice signal by this receiver.
2. speech dialogue system as claimed in claim 1, it is characterized in that, this dialogue control module produces a reply voice signal from this receiver's reply content, carry out specific people's identity by this dialogue control module according to this reply voice signal and confirm flow process, after confirming that this receiver is exactly this specific people, the information of carrying out transmission procedure will send this receiver corresponding to this specific people's information to.
3. speech dialogue system as claimed in claim 1 is characterized in that, also comprises a database, and in order to storing described specific personal data, and one of them passes to this dialogue control module to select described specific personal data, to carry out voice dialogue.
4. speech dialogue system as claimed in claim 2 is characterized in that, this specific people's identity is confirmed included biological attribute data in this specific personal data of flow process basis, judges this reply voice signal of this receiver.
5. speech dialogue system as claimed in claim 4 is characterized in that, this biological attribute data is a sex, data such as age or accent.
6. speech dialogue system as claimed in claim 1 is characterized in that, also comprises and dials and connects module, is connected to this dialogue manager, and according to this specific personal data that this dialogue control module provided, sees through service network and dial and connect connection.
7. speech dialogue system as claimed in claim 6, it is characterized in that this is dialed and connected module and has a phone process analysis (Call Progress Analysis) function, when being communicated with in order to dial and connect, the state of the analysis side talked about, and send this receiver's end state to this dialogue control module.
8. speech dialogue system as claimed in claim 7 is characterized in that, judges that according to the state of this receiver's end whether connection is dialed and connected in this time successful, wherein if this receiver's state is ring unanswered or invalid telephone state, then can directly cuts off and dial and connect.
9. speech dialogue system as claimed in claim 8 is characterized in that, this invalid telephone state comprises spacing, stop words, suspend user mode one of them.
10. speech dialogue system as claimed in claim 8 is characterized in that, this ring unanswered's telephone state comprise in shutdown, unmanned response, the engaged line or in the default stand-by period ring unanswered one of them.
11. speech dialogue system as claimed in claim 7, it is characterized in that, judge according to the state of this receiver's end whether this time connection is successful, if be communicated with successfully, then judge whether to obtain this receiver's dialogic voice data again, if obtain this receiver's dialogic voice data, then carry out this specific people's identity and confirm flow process.
12. speech dialogue system as claimed in claim 7, it is characterized in that, be communicated with successfully if dial and connect, then judge whether to enter voice mail again or recorder is left word, to be subjected to words side be automatic speech exchange or automatic keying exchange if detect this, then imports the voice name or key in extension to reenter dial-up program and phone process analysis and rejudge and dial and connect whether success of connection, and to be subjected to words side be facsimile machine if detect this,, then hang up after the annotation of write-back storehouse.
13. speech dialogue system as claimed in claim 6 is characterized in that, this service network is integrated service digital network (Integrated Service Digital Network).
14. speech dialogue system as claimed in claim 6 is characterized in that, this service network is Public Switched Telephone Network (Public Switched Telephone Network).
15. speech dialogue system as claimed in claim 6 is characterized in that, also comprises a recording module, dials and connects between module and this service network between this, or directly is connected to and dials and connects on the module, in order to the voice dialogue content is recorded.
16. speech dialogue system as claimed in claim 6, it is characterized in that, also comprise a database, in order to store described specific personal data, and one of them passes to this dialogue control module to select described specific personal data, to carry out voice dialogue, in addition in order to store the described voice dialogue content of described specific personal data correspondence.
17. speech dialogue system as claimed in claim 16 is characterized in that, this voice dialogue content comprises this receiver's these dialogic voice data and this reply voice signal, and this inquiry voice signal of exporting of this dialogue manager.
18. speech dialogue system as claimed in claim 17, it is characterized in that, comprise graphical user's interface, be connected to this database and this dials and connects module, in order to obtain the described voice dialogue content of whole or part of described specific personal data correspondence via graphical interface.
19. speech dialogue system as claimed in claim 18 is characterized in that, via this graphical user's interface, directly obtains this receiver's these dialogic voice data and this reply voice signal content.
20. speech dialogue system as claimed in claim 16 is characterized in that, the described voice dialogue content of described storage is in order to improve the foundation of speech recognition as this voice identification module.
21. speech dialogue system as claimed in claim 16 is characterized in that, the described voice dialogue content of described storage is confirmed the operation of flow process in order to improve corresponding described specific personal data with auxiliary this specific people's identity.
22. speech dialogue system as claimed in claim 1, it is characterized in that, also comprise a database, confirm that in order to store this receiver is exactly the time that this specific people answers, and be communicated with the record statistics and estimate make a phone call the time record of (Best Time To Call) of the best according to repeatedly dialing and connecting, touch the rate of reaching with what increase was dialed.
23. speech dialogue system as claimed in claim 1 is characterized in that, this voice identification module is automatic speech recognizing system (Automatic Speech Recognition).
24. speech dialogue system as claimed in claim 1 is characterized in that, this voice generation module is text-to-speech (Text to Speech) module.
25. speech dialogue system as claimed in claim 1 is characterized in that, this inquiry voice signal is to be selected from many voice of pre-recording or produced with synthesis mode by this voice generation module.
26. speech dialogue system as claimed in claim 2, it is characterized in that, when the information of carrying out transmission procedure, in the time of will sending this receiver to corresponding to this specific people's information, further this specific people's of reception vocal print is also compared, if discovery and this specific people's vocal print is not simultaneously, then reenters this specific people's identity and confirm flow process.
27. a speech dialogue system is characterized in that comprising:
One dialogue control module in order to obtaining a specific personal data, and is communicated with a receiver according to this specific personal data; And
One voice generation module, be connected to this dialogue control module, produce an inquiry voice signal corresponding to this specific personal data, and send this receiver to by this dialogue control module, and obtain the voice response signal that this receiver responds this inquiry voice signal, wherein this dialogue control module is carried out specific people's identity affirmation flow process at this receiver's voice response signal, after confirming that this receiver is exactly this specific people, the information of carrying out transmission procedure will send this receiver corresponding to this specific people's information to.
28. speech dialogue system as claimed in claim 27, it is characterized in that, also comprise a voice identification module, be connected to this dialogue control module, in order to receive this voice response signal, produce a voice respective signal after identification, this dialogue control module produces a dialogue control signal according to this voice respective signal, so that produce another inquiry voice signal.
29. speech dialogue system as claimed in claim 27 is characterized in that, also comprises a database, in order to storing described specific personal data, and one of them passes to this dialogue control module to select described specific personal data, to carry out voice dialogue.
30. speech dialogue system as claimed in claim 29 is characterized in that, this specific people's identity is confirmed included biological attribute data in this specific personal data of flow process basis, judges this reply voice signal of this receiver.
31. speech dialogue system as claimed in claim 30 is characterized in that, this biological attribute data is a sex, data such as age or accent.
32. speech dialogue system as claimed in claim 27 is characterized in that, also comprises dialing and connecting module, is connected to this dialogue manager, and according to this specific personal data that this dialogue control module provided, sees through service network and dial and connect connection.
33. speech dialogue system as claimed in claim 32, it is characterized in that this is dialed and connected module and has a phone process analysis (Call Progress Analysis) function, when being communicated with in order to dial and connect, the state of the analysis side talked about, and send this receiver's end state to this dialogue control module.
34. speech dialogue system as claimed in claim 33 is characterized in that, judges that according to the state of this receiver's end whether connection is dialed and connected in this time successful, wherein if this receiver's state is ring unanswered or invalid telephone state, then can directly cuts off and dial and connect.
35. speech dialogue system as claimed in claim 34 is characterized in that, this invalid telephone state comprises spacing, stop words, suspend user mode one of them.
36. speech dialogue system as claimed in claim 34 is characterized in that, this ring unanswered's telephone state comprise in shutdown, unmanned response, the engaged line or in the default stand-by period ring unanswered one of them.
37. speech dialogue system as claimed in claim 32, it is characterized in that, be communicated with successfully if dial and connect, then judge whether to enter voice mail again or recorder is left word, if detecting this, to be subjected to words side be automatic speech exchange or automatic keying exchange, then import the voice name or key in extension and reenter dial-up program and phone process analysis and rejudge and dial and connect whether success of connection, to be subjected to words side be facsimile machine if detect this, then hangs up after the annotation of write-back storehouse.
38. speech dialogue system as claimed in claim 32 is characterized in that, this service network is integrated service digital network (Integrated Service Digital Network).
39. speech dialogue system as claimed in claim 32 is characterized in that, this service network is Public Switched Telephone Network (Public Switched Telephone Network).
40. speech dialogue system as claimed in claim 32 is characterized in that, also comprises a recording module, dials and connects between module and this service network between this, or directly is connected to and dials and connects on the module, in order to the voice dialogue content is recorded.
41. speech dialogue system as claimed in claim 32, it is characterized in that, also comprise a database, in order to store described specific personal data, and one of them passes to this dialogue control module to select described specific personal data, to carry out voice dialogue, in addition in order to store the described voice dialogue content of described specific personal data correspondence.
42. speech dialogue system as claimed in claim 32, it is characterized in that, also comprise graphical user's interface, be connected to this database and this dials and connects module, in order to obtain the described voice dialogue content of whole or part of described specific personal data correspondence via graphical interface.
43. speech dialogue system as claimed in claim 37, it is characterized in that, also comprise a database, confirm that in order to store this receiver is exactly the time that this specific people answers, and be communicated with the record statistics and estimate make a phone call the time record of (Best Time To Call) of the best according to repeatedly dialing and connecting, touch the rate of reaching with what increase was dialed.
44. speech dialogue system as claimed in claim 37, it is characterized in that, when the information of carrying out transmission procedure, in the time of will sending this receiver to corresponding to this specific people's information, further this specific people's of reception vocal print is also compared, if discovery and this specific people's vocal print is not simultaneously, then reenters this specific people's identity and confirm flow process.
45. speech dialogue system as claimed in claim 27 is characterized in that, this inquiry voice signal is by selecting from many voice of pre-recording or being synthesized generation.
46. a voice dialogue processing method is characterized in that comprising:
Obtain a specific personal data, inquire voice signal according to this specific personal data to produce one first, and according to this specific personal data with after a receiver is communicated with, send this first inquiry voice signal to this receiver, wherein, reply the back by this receiver according to this first inquiry voice signal and produce a reply voice signal;
Behind this reply voice signal of identification, obtain a voice respective signal, carry out this specific people's identity according to this voice respective signal and confirm flow process, and confirm that according to this specific people's identity the result of flow process produces a dialogue control signal, and produce a voice responsive signal according to this, and send the receiver to and continue this specific people's identity and confirm or end dialog.
47. voice dialogue processing method as claimed in claim 46 is characterized in that, this specific people's identity is carried out the information transmission procedure after confirming that flow process confirms that this answers artificial this specific people, will send this receiver corresponding to this specific people's information to.
48. voice dialogue processing method as claimed in claim 46 is characterized in that, this specific people's identity is confirmed included biological attribute data in this specific personal data of flow process basis, judges this receiver's reply voice signal.
49. voice dialogue processing method as claimed in claim 48 is characterized in that, this biological attribute data is sex, age or accent data.
50. voice dialogue processing method as claimed in claim 46, it is characterized in that, this mode of communicating is for to carry out via the module of dialing and connecting with phone process analysis (Call Progress Analysis) function, and dialing and connecting when being communicated with, and analyzes and obtain the state of receiver's end.
51. voice dialogue processing method as claimed in claim 50 is characterized in that, judges that according to the state of this receiver's end whether this time connection is successful, when wherein the state of holding as if this receiver is ring unanswered or invalid phone, then can directly cuts off and dial and connect.
52. voice dialogue processing method as claimed in claim 50, it is characterized in that, judge according to the state of this receiver's end whether connection is dialed and connected in this time successful, wherein if the state of this receiver's end is a valid telephone, then judge whether the dialogic voice data that enter voice mail or obtain this receiver again, if obtain this receiver's dialogic voice data, then carry out this specific people's identity and confirm flow process.
53. voice dialogue processing method as claimed in claim 50 is characterized in that, if be communicated with successfully, judges whether that then entering voice mail leaves word again.
54. voice dialogue processing method as claimed in claim 46, it is characterized in that, comprise the described voice dialogue content that stores described receiver's correspondence, wherein this voice dialogue content comprises this receiver's these dialogic voice data and this reply voice signal, and this first inquiry voice signal of output.
55. voice dialogue processing method as claimed in claim 54 is characterized in that, the described voice dialogue content of described storage is in order to revise the foundation of speech recognition as this voice identification module.
56. voice dialogue processing method as claimed in claim 54 is characterized in that, the described voice dialogue content of described storage is confirmed the operation of flow process in order to revise corresponding described specific personal data with auxiliary this specific people's identity.
57. voice dialogue processing method as claimed in claim 46, it is characterized in that, also comprise storing and confirm that this receiver is exactly the time that this specific people answers, and be communicated with record and obtain make a phone call the time record of (Best Time To Call) of the best according to repeatedly dialing and connecting, touch the rate of reaching with what increase was dialed.
58. voice dialogue processing method as claimed in claim 46, it is characterized in that, comprise that also the result who obtains for connection confirms that with this specific people's identity the result that flow process obtains produces corresponding behavior code (Action Code), and stored, and as dialogue controlling mechanism correction foundation.
59. voice dialogue processing method as claimed in claim 46, it is characterized in that, when the information of carrying out transmission procedure, in the time of will sending this receiver to corresponding to this specific people's information, further detect unusual condition, if when finding that this unusual condition takes place, then reenter this specific people's identity and confirm flow process.
60. voice dialogue processing method as claimed in claim 59 is characterized in that, this unusual condition is not then assert this unusual condition simultaneously for the vocal print of finding this receiver and is taken place.
61. voice dialogue processing method as claimed in claim 59 is characterized in that, this unusual condition can't assert that then this unusual condition takes place during identification for the answer of finding this receiver continues a set number of times.
62. voice dialogue processing method as claimed in claim 59 is characterized in that, this unusual condition takes place for this unusual condition of then assert when a period of time does not have any response of finding this receiver.
63. voice dialogue processing method as claimed in claim 59, it is characterized in that, after reentering this specific people's identity affirmation flow process, inquire this receiver according to this specific personal data generation and the unduplicated second inquiry voice signal of this first inquiry voice signal.
64. voice dialogue processing method as claimed in claim 59 is characterized in that, remains unfulfilled affirmation after this second inquiry voice signal is inquired this receiver, then carrying out other unduplicated inquiry voice signal, up to finishing affirmation.
65. voice dialogue processing method as claimed in claim 46, it is characterized in that, when also not transmit, do not finish this inquiry voice signal, when this receiver just hangs up the telephone, then can judge the whether complete information that receives transmission of this receiver, and give different behavior codes according to meaning of one's words halt.
66. speech dialogue system as claimed in claim 46 is characterized in that, this inquiry voice signal is by selecting from many voice of pre-recording or producing with synthesis mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100955873A CN101572756A (en) | 2008-04-29 | 2008-04-29 | Dialogue system and voice dialogue processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100955873A CN101572756A (en) | 2008-04-29 | 2008-04-29 | Dialogue system and voice dialogue processing method |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410814622.8A Division CN104601832A (en) | 2008-04-29 | 2008-04-29 | Conversation system and speech conversation processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101572756A true CN101572756A (en) | 2009-11-04 |
Family
ID=41231985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008100955873A Pending CN101572756A (en) | 2008-04-29 | 2008-04-29 | Dialogue system and voice dialogue processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101572756A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104183238A (en) * | 2014-09-09 | 2014-12-03 | 深圳市和音达科技有限公司 | Old people voiceprint recognition method based on questioning and answering |
| CN105426675A (en) * | 2015-11-13 | 2016-03-23 | 江苏大学 | Full-automatic hospital telephone follow-up method and telephone device thereof |
| CN105554330A (en) * | 2016-01-06 | 2016-05-04 | 努比亚技术有限公司 | Voice message device and method |
| CN109120804A (en) * | 2017-06-26 | 2019-01-01 | 浙江金讯网络技术股份有限公司 | A kind of outer paging system based on response message identification |
| CN109660678A (en) * | 2018-12-07 | 2019-04-19 | 深圳前海微众银行股份有限公司 | Electric core network system realization, system and readable storage medium storing program for executing |
| CN109873909A (en) * | 2019-03-15 | 2019-06-11 | 腾讯科技（深圳）有限公司 | A kind of voice call method, device and equipment and computer storage medium |
| WO2019153346A1 (en) * | 2018-02-12 | 2019-08-15 | 深圳迎凯生物科技有限公司 | In-vitro diagnosis device control method and device, computer device and storage medium |
| CN110265023A (en) * | 2019-05-20 | 2019-09-20 | 南京觉醒智能科技有限公司 | A kind of both sides scene words art real-time online quality detecting method |
| WO2021047319A1 (en) * | 2019-09-11 | 2021-03-18 | 深圳壹账通智能科技有限公司 | Voice-based personal credit assessment method and apparatus, terminal and storage medium |
-
2008
- 2008-04-29 CN CNA2008100955873A patent/CN101572756A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104183238A (en) * | 2014-09-09 | 2014-12-03 | 深圳市和音达科技有限公司 | Old people voiceprint recognition method based on questioning and answering |
| CN105426675A (en) * | 2015-11-13 | 2016-03-23 | 江苏大学 | Full-automatic hospital telephone follow-up method and telephone device thereof |
| CN105554330A (en) * | 2016-01-06 | 2016-05-04 | 努比亚技术有限公司 | Voice message device and method |
| CN109120804A (en) * | 2017-06-26 | 2019-01-01 | 浙江金讯网络技术股份有限公司 | A kind of outer paging system based on response message identification |
| WO2019153346A1 (en) * | 2018-02-12 | 2019-08-15 | 深圳迎凯生物科技有限公司 | In-vitro diagnosis device control method and device, computer device and storage medium |
| CN109660678A (en) * | 2018-12-07 | 2019-04-19 | 深圳前海微众银行股份有限公司 | Electric core network system realization, system and readable storage medium storing program for executing |
| CN109873909A (en) * | 2019-03-15 | 2019-06-11 | 腾讯科技（深圳）有限公司 | A kind of voice call method, device and equipment and computer storage medium |
| CN110265023A (en) * | 2019-05-20 | 2019-09-20 | 南京觉醒智能科技有限公司 | A kind of both sides scene words art real-time online quality detecting method |
| WO2021047319A1 (en) * | 2019-09-11 | 2021-03-18 | 深圳壹账通智能科技有限公司 | Voice-based personal credit assessment method and apparatus, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101572756A (en) | Dialogue system and voice dialogue processing method | |
| TWI521936B (en) | Outbound dialogue system and dialogue operation method | |
| CN109413286B (en) | Intelligent customer service voice response system and method | |
| US5953393A (en) | Personal telephone agent | |
| CN109672786B (en) | Incoming call answering method and device | |
| CN110891124A (en) | System for artificial intelligence pick-up call | |
| Arons | The Audio-Graphical Interface to a Personal Integrated Telecommunications System | |
| US6792082B1 (en) | Voice mail system with personal assistant provisioning | |
| CN101128811B (en) | Method, system and service for achieving synchronous communication responsive to dynamic status | |
| US8983051B2 (en) | Outgoing call classification and disposition | |
| EP0935378B1 (en) | System and methods for automatic call and data transfer processing | |
| EP1768369A1 (en) | Presence and preference enabled voice response system and method | |
| CN201504266U (en) | User voice processing system based on telephone bank | |
| US20070223662A1 (en) | Content sensitive do-not-disturb (dnd) | |
| US20090104898A1 (en) | A telephone using a connection network for processing data remotely from the telephone | |
| CN105979107B (en) | Automatic outer call method and system based on speech recognition | |
| KR102136706B1 (en) | Information processing system, reception server, information processing method and program | |
| CN103295585A (en) | Processing system and method identifying whether telephone automatic dialing is responded by real human | |
| CN101141530A (en) | Centralised storage of data | |
| US6563911B2 (en) | Speech enabled, automatic telephone dialer using names, including seamless interface with computer-based address book programs | |
| CN104601832A (en) | Conversation system and speech conversation processing method | |
| US20020118803A1 (en) | Speech enabled, automatic telephone dialer using names, including seamless interface with computer-based address book programs, for telephones without private branch exchanges | |
| EP1511277A1 (en) | Method for answering an incoming event with a phone device, and adapted phone device | |
| CN103634448A (en) | Method for intelligently responding to incoming calls by voice | |
| CN113779217A (en) | Intelligent voice outbound service method and system based on human-computer interaction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20091104 |