CN109227558A - Can in real time adjustment intelligent outgoing call robot - Google Patents
Can in real time adjustment intelligent outgoing call robot Download PDFInfo
- Publication number
- CN109227558A CN109227558A CN201811174233.8A CN201811174233A CN109227558A CN 109227558 A CN109227558 A CN 109227558A CN 201811174233 A CN201811174233 A CN 201811174233A CN 109227558 A CN109227558 A CN 109227558A
- Authority
- CN
- China
- Prior art keywords
- robot
- client
- outgoing call
- development
- reply
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/0005—Manipulators having means for high-level communication with users, e.g. speech generator, face recognition means
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses can in real time adjustment intelligent outgoing call robot, development/testing stage and online stage, the development/testing stage initiates dialogue by the outside pager device people of phone by development/testing personnel, and the reply of outgoing call robot then passes through text and is showed on a Debugging interface, the online stage is that all interaction is completed by phone for client and outgoing call robot.Present invention reduces the difficulty of configuration, the effect tested in configuration can reach, while can also adjust in real time script for story-telling according to the actual effect after online, reduce the cost of artificial discs.
Description
Technical field
The present invention relates to outgoing call robotic technology fields, more particularly to can in real time adjustment intelligent outgoing call robot.
Background technique
Robot (Robot) is the automatic installations for executing work.It can not only receive mankind commander, but also can run
The program of preparatory layout, can also be according to principle program action formulated with artificial intelligence technology.Its task is to assist or take
For the work of human work, such as production industry, construction industry, or dangerous work.
The development of current information technology high-speed, this has also just driven flourishing for Robot industry, has exhaled robot field outside
Jing Zhong, traditional outgoing call robot mainly by the way of rule configuration, according to the script for story-telling script that client provides, pass through keyword
Or the modes such as regular expression write the corresponding rule of each script for story-telling node in script, record after online further according to practical outgoing call
Sound carries out discs and carries out the adjustment of script for story-telling or rule again, however has the disadvantage in that rule writes difficulty in this way, it is difficult to cover
Various unknown situations；Iteration adjustment speed is slow, and discs needs plenty of time energy.Thus we have proposed can in real time adjustment intelligence
It can outgoing call robot.
Summary of the invention
The purpose of the present invention is to solve disadvantage existing in the prior art, and propose can adjustment in real time intelligence it is outer
Exhale robot.
To achieve the goals above, present invention employs following technical solutions:
Can adjustment in real time intelligent outgoing call robot, including development/testing stage and online stage, the development/testing
Stage initiates dialogue by the outside pager device people of phone by development/testing personnel, and the reply of outgoing call robot then passes through text
It is showed on a Debugging interface, the development/testing stage includes following below scheme:
A. development/testing personnel talk, and identification is intended to and provides reply after robot receives；
B. the reply of robot meets expection, is directly entered step e, otherwise enters step c；
C. development/testing personnel input instruction notification robot, and last round of intention assessment is incorrect, while inputting correct
Intention number；
D. after robot receives correct intent instructions, by automatically generating new rule/model incremental learning/extensive chemical
The modes such as habit are adapted to；
E. continue next round test；
The online stage is that all interaction is completed by phone for client and outgoing call robot, the online stage packet
Include following below scheme:
A. client talks, and robot identification client provides reply after being intended to；
B. client continues words after the reply for hearing robot；
Whether it is the direct response spoken to last round of robot that C. algorithm judges that client responds, if it is, into D ring
Section, but regardless of whether be directly to respond, outgoing call robot requires to continue conversation process；
D. algorithm judges whether the response of client is actively to respond, and responds if it is positive, then the reply of identified machine people
It is a positive sample under this node, is otherwise labeled as a negative sample；
E. by way of on-line study, the effect of this node intention assessment of adjust automatically.
Preferably, the definition actively responded includes two o'clock, and first point: robot initiates to put question to, and client is gone wrong
Answer, robot provides suggestion, and client receives suggestion, second point: robot is answered the problem of client, and client indicates to answering
It is satisfied.
Compared with prior art, the beneficial effects of the present invention are: the present invention is by introducing a kind of Real-time Feedback learning machine
System, the adjustment that can be dynamically configured during calling work as chance when robot being exhaled to enter the development/testing stage outside
To not yet capped topic, to produce when not meeting expected robot and replying, tester can pass through default refer to
Order feeds back to robot, allows robot adjust automatically matching algorithm to learn the unknown script for story-telling, meanwhile, robot is being exhaled outside just
After formula is online, according to the reply content of client, to judge whether robot is correct in last round of reply, and according to this dynamic
The script for story-telling of robot is adjusted, which reduces the difficulty of configuration, can reach the effect tested in configuration, while can be with root
Script for story-telling is adjusted in real time according to the actual effect after online, reduces the cost of artificial discs.
Detailed description of the invention
Fig. 1 be it is proposed by the present invention can in real time adjustment intelligent outgoing call robot development/testing stage and online stage
Flowage structure schematic diagram.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.
Referring to Fig.1, can adjustment in real time intelligent outgoing call robot, it is including development/testing stage and online stage, described to open
Hair/test phase initiates dialogue by the outside pager device people of phone by development/testing personnel, and the reply of outgoing call robot is then led to
It crosses text to be showed on a Debugging interface, the development/testing stage includes following below scheme:
A. development/testing personnel talk, and identification is intended to and provides reply after robot receives；
B. the reply of robot meets expection, is directly entered step e, otherwise enters step c；
C. development/testing personnel input instruction notification robot, and last round of intention assessment is incorrect, while inputting correct
Intention number；
D. after robot receives correct intent instructions, by automatically generating new rule/model incremental learning/extensive chemical
The modes such as habit are adapted to；
E. continue next round test；
The online stage is that all interaction is completed by phone for client and outgoing call robot, the online stage packet
Include following below scheme:
A. client talks, and robot identification client provides reply after being intended to；
B. client continues words after the reply for hearing robot；
Whether it is the direct response spoken to last round of robot that C. algorithm judges that client responds, if it is, into D ring
Section, but regardless of whether be directly to respond, outgoing call robot requires to continue conversation process；
D. algorithm judges whether the response of client is actively to respond, and responds if it is positive, then the reply of identified machine people
It is a positive sample under this node, is otherwise labeled as a negative sample；
E. by way of on-line study, the effect of this node intention assessment of adjust automatically.
The definition actively responded includes two o'clock, and first point: robot initiates to put question to, and client gives the answer gone wrong, machine
People provides suggestion, and client receives suggestion, second point: the problem of client, is answered by robot, and client is satisfied with to answer.
In the present embodiment, firstly, the present invention is by introducing a kind of Real-time Feedback study mechanism, it can be during calling
The adjustment that is configured of dynamic, when robot being exhaled to enter the development/testing stage outside, when encountering not yet capped topic, from
And produce when not meeting expected robot and replying, tester can feed back to robot by preset instructions, allow machine
People's adjust automatically matching algorithm to learning the unknown script for story-telling, meanwhile, exhale outside robot it is formal it is online after, according to client's
Reply content, to judge whether robot is correct in last round of reply, and according to the script for story-telling of this dynamic adjusting machine device people, the hair
The bright difficulty for reducing configuration, can reach effect test in configuration, while can also be according to the actual effect reality after online
When adjust script for story-telling, reduce the cost of artificial discs.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art in the technical scope disclosed by the present invention, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (2)
1. can adjustment in real time intelligent outgoing call robot, including development/testing stage and online stage, which is characterized in that described
The development/testing stage initiates dialogue by the outside pager device people of phone by development/testing personnel, and the reply of outgoing call robot is then
It is showed on a Debugging interface by text, the development/testing stage includes following below scheme:
A. development/testing personnel talk, and identification is intended to and provides reply after robot receives；
B. the reply of robot meets expection, is directly entered step e, otherwise enters step c；
C. development/testing personnel input instruction notification robot, and last round of intention assessment is incorrect, while inputting correctly meaning
Figure number；
D. after robot receives correct intent instructions, by automatically generating new rule/model incremental learning/intensified learning etc.
Mode is adapted to；
E. continue next round test；
The online stage is that all interactions are completed by phone for client and outgoing call robot, the online stage include with
Lower process:
A. client talks, and robot identification client provides reply after being intended to；
B. client continues words after the reply for hearing robot；
Whether it is the direct response spoken to last round of robot that C. algorithm judges that client responds, if it is, into D link,
But regardless of whether being directly to respond, outgoing call robot requires to continue conversation process；
D. algorithm judges whether the response of client is actively to respond, and responds if it is positive, then the reply of identified machine people is this
A positive sample under a node, is otherwise labeled as a negative sample；
E. by way of on-line study, the effect of this node intention assessment of adjust automatically.
2. it is according to claim 1 can adjustment in real time intelligent outgoing call robot, which is characterized in that it is described actively to respond
Definition includes two o'clock, and first point: robot initiates to put question to, and client gives the answer gone wrong, and robot provides suggestion, and client receives
It is recommended that second point: the problem of client, is answered by robot, and client is satisfied with to answer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811174233.8A CN109227558A (en) | 2018-10-09 | 2018-10-09 | Can in real time adjustment intelligent outgoing call robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811174233.8A CN109227558A (en) | 2018-10-09 | 2018-10-09 | Can in real time adjustment intelligent outgoing call robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109227558A true CN109227558A (en) | 2019-01-18 |
Family
ID=65054356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811174233.8A Pending CN109227558A (en) | 2018-10-09 | 2018-10-09 | Can in real time adjustment intelligent outgoing call robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109227558A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111464701A (en) * | 2020-03-12 | 2020-07-28 | 云知声智能科技股份有限公司 | Method and device for carrying out simulation multi-round test on robot outbound telephone system |
| CN111696576A (en) * | 2020-05-21 | 2020-09-22 | 升智信息科技(南京)有限公司 | Intelligent voice robot talk test system |
| CN113448871A (en) * | 2021-07-22 | 2021-09-28 | 深圳追一科技有限公司 | Session debugging method and device, computer equipment and computer-readable storage medium |
| WO2022089546A1 (en) * | 2020-10-28 | 2022-05-05 | 华为云计算技术有限公司 | Label generation method and apparatus, and related device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1614630A (en) * | 2004-11-29 | 2005-05-11 | 南京大学 | Rapid study classifying method |
| CN106897263A (en) * | 2016-12-29 | 2017-06-27 | 北京光年无限科技有限公司 | Robot dialogue exchange method and device based on deep learning |
| CN107357838A (en) * | 2017-06-23 | 2017-11-17 | 上海交通大学 | Dialog strategy canbe used on line method based on multi-task learning |
| CN107851434A (en) * | 2015-05-26 | 2018-03-27 | 鲁汶大学 | Use the speech recognition system and method for auto-adaptive increment learning method |
| US20180114125A1 (en) * | 2016-10-24 | 2018-04-26 | Fuji Xerox Co., Ltd. | Emotion estimating system |
| CN108363690A (en) * | 2018-02-08 | 2018-08-03 | 北京十三科技有限公司 | Dialog semantics Intention Anticipation method based on neural network and learning training method |
-
2018
- 2018-10-09 CN CN201811174233.8A patent/CN109227558A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1614630A (en) * | 2004-11-29 | 2005-05-11 | 南京大学 | Rapid study classifying method |
| CN107851434A (en) * | 2015-05-26 | 2018-03-27 | 鲁汶大学 | Use the speech recognition system and method for auto-adaptive increment learning method |
| US20180114125A1 (en) * | 2016-10-24 | 2018-04-26 | Fuji Xerox Co., Ltd. | Emotion estimating system |
| CN106897263A (en) * | 2016-12-29 | 2017-06-27 | 北京光年无限科技有限公司 | Robot dialogue exchange method and device based on deep learning |
| CN107357838A (en) * | 2017-06-23 | 2017-11-17 | 上海交通大学 | Dialog strategy canbe used on line method based on multi-task learning |
| CN108363690A (en) * | 2018-02-08 | 2018-08-03 | 北京十三科技有限公司 | Dialog semantics Intention Anticipation method based on neural network and learning training method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111464701A (en) * | 2020-03-12 | 2020-07-28 | 云知声智能科技股份有限公司 | Method and device for carrying out simulation multi-round test on robot outbound telephone system |
| CN111696576A (en) * | 2020-05-21 | 2020-09-22 | 升智信息科技(南京)有限公司 | Intelligent voice robot talk test system |
| WO2022089546A1 (en) * | 2020-10-28 | 2022-05-05 | 华为云计算技术有限公司 | Label generation method and apparatus, and related device |
| CN113448871A (en) * | 2021-07-22 | 2021-09-28 | 深圳追一科技有限公司 | Session debugging method and device, computer equipment and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109227558A (en) | Can in real time adjustment intelligent outgoing call robot | |
| Pardos et al. | Learning gain differences between ChatGPT and human tutor generated algebra hints | |
| Winkler et al. | Alexa, can you help us solve this problem? How conversations with smart personal assistant tutors increase task group outcomes | |
| Gorman et al. | What's learned together stays together: speakers' choice of referring expression reflects shared experience. | |
| Dewaele et al. | The link between foreign language classroom anxiety, second language tolerance of ambiguity and self-rated English proficiency among Chinese learners | |
| Yoon et al. | Adjusting conceptual pacts in three-party conversation. | |
| Utami et al. | Collaborative user responses in multiparty interaction with a couples counselor robot | |
| Zhexenova et al. | A comparison of social robot to tablet and teacher in a new script learning context | |
| CN110489756A (en) | Conversational human-computer interaction spoken language evaluation system | |
| Hung et al. | Towards a method for evaluating naturalness in conversational dialog systems | |
| Tsui et al. | Impact of language dominance on phonetic transfer in Cantonese–English bilingual language switching | |
| CN106127526A (en) | Intelligent robot system and method for work thereof | |
| Khouzaimi et al. | Optimising turn-taking strategies with reinforcement learning | |
| Sinclair et al. | Tutorbot Corpus: Evidence of Human-Agent Verbal Alignment in Second Language Learner Dialogues. | |
| El-Mansy et al. | Investigating small-group cognitive engagement in general chemistry learning activities using qualitative content analysis and the ICAP framework | |
| CN109859559A (en) | A kind of substation safety holds rule emulation training method | |
| Stroud | The concept of semilingualism | |
| Briggs et al. | When robots object: Evidence for the utility of verbal, but not necessarily spoken protest | |
| Nath | Towards naturally responsive spoken dialog systems by modelling pragmatic-prosody correlations of discourse markers | |
| Malchanau et al. | Multidimensional dialogue management for tutoring systems | |
| Zhang et al. | GrounDialog: A Dataset for Repair and Grounding in Task-oriented Spoken Dialogues for Language Learning | |
| Ariyoshi et al. | Effect of facial colors on humanoids in emotion recognition using speech | |
| Kurashige et al. | Counseling robot implementation and evaluation | |
| Danilava et al. | Why are you silent?-Towards responsiveness in chatbots | |
| Low et al. | Implementation of a Virtual Patient Chatbot for Physiotherapy Students Training |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |