CN109151220A - A kind of communication session call failure scene analysis system - Google Patents
A kind of communication session call failure scene analysis system Download PDFInfo
- Publication number
- CN109151220A CN109151220A CN201811054509.9A CN201811054509A CN109151220A CN 109151220 A CN109151220 A CN 109151220A CN 201811054509 A CN201811054509 A CN 201811054509A CN 109151220 A CN109151220 A CN 109151220A
- Authority
- CN
- China
- Prior art keywords
- state
- keyword
- module
- call
- called
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/22—Arrangements for supervision, monitoring or testing
- H04M3/2272—Subscriber line supervision circuits, e.g. call detection circuits
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/42008—Systems for anonymous communication between parties, e.g. by use of disposal contact identifiers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
- H04M3/4936—Speech interaction details
Abstract
The invention discloses a kind of communication session call failure scene analysis systems, belong to communication service technical field.Traffic platform, audio storage module, keyword recognition module and state decision discernment module including safety call center;Calling subscriber initiates dial-up to called subscriber by the intermediary number of the traffic platform;The traffic platform carries out speech admission to calling called subscriber's process, and after being determined as call failure, the speech of failure is stored to audio storage module;The keyword recognition module identifies the voice audio of failure, and exports keyword;After the state decision discernment module is by analyzing the keyword of identification, the called state that can not call connection of output.The present invention is closed loop feedback, its own can constantly improve system, and enable a system to identification full dose is not turned on scene.
Description
Technical field
The present invention relates to communication service technical field, especially a kind of communication session call failure scene analysis system.
Background technique
Safety call center is to provide the platform of mobile phone safe secret protection, client's kimonos for O2O Internet enterprises user
Business personnel reach outer aobvious intermediary number each other by safety call center in communication process, hide the purpose of true number.
Include the scene of many call failures in voice call process, is divided into spacing, called subscriber's shutdown, called does not exist
Service area, caller defaulting subscriber/partial suspension, user apply for hesitation/other reasons hesitation, called shutdown, are called
Busy (Call Waiting and dial again later), called subscriber be provided with incoming call limitation, international long-distance limitation, plus trunk code, other
Operator's direction line failure, long-distance limitation/illegal caller, called long time no answer, the accurate positioning of these scenes is to third party
There are very high data mining commercial values for the internet O2O.
Existing analysis system is usually to be solved by signaling processing, is determined by the signaling code returned after call failure
The reason of call failure, and then scene classification is carried out to call, and there is no a unified required standards to come between operator
Constrain the unified relationship of called state and signaling code, and the signaling scheme that uses of some telephone networks for call failure playback after
Return to signaling, if during playback caller hang up if return to the signaling code that caller hangs up, leading to not to obtain called can not connect
Logical real scene state.
Fourier transform is based in paper " speech recognition technology heart automatic outer call is applied in a call research and realization "
Comparing with sample for characteristic value is extracted, ringing tone state recognition is realized, but since its principle is sample comparison, to some more multiple
Miscellaneous scene then can not Accurate Analysis positioning, for example be that prompt other side is busy now when putting through, ask should not on-hook, called
When being responsible for hanging up calling, then other side is prompted to be inconvenient to answer, in paper " speech recognition technology heart automatic outer call application in a call
Research and realization " in can not to the complex scene carry out Accurate Analysis positioning.
This system is not turned on state and identified using data stream architecture to called, is first extracted by speech recognition technology
Keyword, decision tree branches algorithm identifies keyword by setting each node data, may be implemented each to called state
Kind complex scene is precisely identified.
Summary of the invention
Goal of the invention of the invention is, in view of the above-mentioned problems, a kind of communication session call failure scene analysis system is provided,
Identification classification can be carried out to the scene of the non-answered call of the internet third party O2O, to further data mining, value
It was found that etc. analyses operation provide data supporting.
In order to achieve the above objectives, the technical scheme adopted by the invention is that:
A kind of communication session call failure scene analysis system, it is characterised in that: the traffic including safety call center is flat
Platform, audio storage module, keyword recognition module and state decision discernment module;
Calling subscriber initiates dial-up to called subscriber by the intermediary number of the traffic platform;
The traffic platform carries out speech admission to calling called subscriber's process, after being determined as call failure, by failure
Speech is stored to audio storage module;
The keyword recognition module identifies the voice audio of failure, and exports keyword;
After the state decision discernment module is by analyzing the keyword of identification, output is called can not to call connection
State.
Preferably, it in dial-up, after media negotiation, records to called playback, when called state playback is normal
When terminating and returning to signaling code, recording is abandoned, records called state;At the end of non-normal playback, to by audio sound-recording store to
Audio storage module.
Preferably, after the keyword recognition module obtains the audio sound-recording of audio storage module, identification process with it is described
State decision discernment module is interacted to recording end of identification.
Preferably, the state decision discernment module is using the tree-like branch detection structure of stake is determined, according to the field of call failure
Scape, i.e. spacing, called subscriber's shutdown, called out of reach, caller defaulting subscriber/partial suspension, user apply hesitation/
Other reasons hesitation, called shutdowns, Called Busy, Call Waiting, dial again later, called subscriber be provided with incoming call limit, state
The long-distance limitation in border does not add trunk code, other operator's direction line failures, long-distance limitation/illegal caller, called long time no answer
State carries out corresponding keyword extraction.
Preferably, the state recognition keyword root includes: " you are good ", " sorry ", " number ", " electricity according to state extraction
Words ", on-hook, " user ", " dialing ", " network ", " spacing ", " trunk code ", " shutdown ", " in call ", " not yet ", " state
Border ", " in call ", " being not turned on ", " incoming call limitation ", " arrearage ", " shutdown ", " service ", " ring unanswered ", " shutdown " are closed
It is multiplexed and is associated with each other between key word, determine to may recognize that called state through most four times.
Preferably, the keyword recognition module and the state decision discernment module interactive process, the speech recognition
Module carries out keyword recognition according to the keyword that pre-sets, can not be in identified situation in voice, and voice is identified
And it carries out artificial intervention analysis and is not turned on state, and extract keyword and be supplemented in state decision discernment module.
Preferably, in the keyword recognition module detection process, the identical list of keywords of automatic shield, it is subsequent again
The continuous keyword of the series then automatic shield is recognized, guarantee is transferred to state recognition module without repetition list of keywords.
Preferably, the keyword recognition module carries out keyword to carry out one to all keywords in identification process
It is secondary to re-enter detection, it detects whether as " you are good " and " sorry ", if front and back detects that state is inconsistent twice, according to suitable
Sequence exports successive identification state.
Due to the adoption of the above technical scheme, the invention has the following advantages:
1. the present invention is closed loop feedback, its own can constantly improve system, enable a system to can not connecing for identification full dose
Logical scene.Dress light weight detection identification can be determined simultaneously, shorten recognition cycle.
2. the present invention is able to detect that the operation of user's subjectivity and objective situation bring multimode by cycle detection
Comprehensive scene.Fourier transform is based in paper " speech recognition technology heart automatic outer call is applied in a call research and realization "
Comparing with sample for characteristic value is extracted, ringing tone state recognition is realized, but since its principle is sample comparison, to some more multiple
Miscellaneous scene then can not Accurate Analysis positioning, for example be that prompt other side is busy now when putting through, ask should not on-hook, called
When being responsible for hanging up calling, then other side is prompted to be inconvenient to answer, in paper " speech recognition technology heart automatic outer call application in a call
Research and realization " in can not to the complex scene carry out Accurate Analysis positioning.This system is using data stream architecture to called nothing
Method on-state is identified, first extracts keyword by speech recognition technology, decision tree branches algorithm is by setting each node
Data identify keyword, may be implemented precisely to identify the various complex scenes of called state.
Detailed description of the invention
Fig. 1 is present system structural block diagram.
Fig. 2 is the present invention reply normal scene principal diagram of playback.
Fig. 3 is that scene principal diagram is not finished in present invention reply playback.
Fig. 4 is keyword identification of the present invention and states detecting theory schematic diagram.
Specific embodiment
It is further illustrated below in conjunction with specific implementation of the attached drawing to invention.
As shown in Figure 1, the present invention includes traffic platform, audio storage module, the keyword recognition mould at safety call center
Block and state decision discernment module.Wherein, traffic platform is connected with speech path network, is used for incoming call and exhalation, traffic platform and sound
Frequency memory module is connected, and actively pushes recording file to audio storage module;Audio storage module and keyword recognition module phase
Even, think that keyword recognition module pushes audio sound-recording;Keyword recognition module is connected with state decision discernment module backward, is used for
To the keyword of state decision discernment module push identification.Its specifically:
Calling subscriber initiates dial-up to called subscriber by the intermediary number of the traffic platform;
The traffic platform carries out speech admission to calling called subscriber's process, after being determined as call failure, by failure
Speech is stored to audio storage module;
The keyword recognition module identifies the voice audio of failure, and exports keyword;
After the state decision discernment module is by analyzing the keyword of identification, output is called can not to call connection
State.
As Figure 2-3, it in dial-up, after media negotiation, records to called playback, when called state playback
Normal termination and when returning to signaling code, abandons recording, records called state.At the end of non-normal playback, deposited to by audio sound-recording
It stores up to audio storage module.After keyword recognition module obtains the audio sound-recording of audio storage module, identification process and the shape
State decision discernment module is interacted to recording end of identification.
In the present system, state decision discernment module is using the tree-like branch detection structure of stake is determined, according to the field of call failure
Scape, i.e. spacing, called subscriber's shutdown, called out of reach, caller defaulting subscriber/partial suspension, user apply hesitation/
Other reasons hesitation, called shutdowns, Called Busy, Call Waiting, dial again later, called subscriber be provided with incoming call limit, state
The long-distance limitation in border does not add trunk code, other operator's direction line failures, long-distance limitation/illegal caller, called long time no answer
State carries out corresponding keyword extraction.Specifically, including " you are good " and " sorry " two-way branch, in the lower layer point of two-way branch
Multiple layers of branch are not set, such as " number ", " phone ", on-hook, " user " etc. are set in " you are good " branch subordinate;" number
" spacing ", " trunk code " is arranged in code " subordinate;In " phone " subordinate, " shutdown ", " in call ", " not yet " are set;In on-hook
" phone " is arranged in subordinate;At " user " subordinate setting " just busy ".In " sorry " branch subordinate, " user ", " dialing ", " electricity are set
Words ", " network " etc.;In " dialing " subordinate, " shutdown " is set;In " phone " subordinate, " arrearage ", " shutdown ", " not yet ", " nothing are set
People answers ";In " network " subordinate, " being not turned on " etc. is set.Here next layer of keyword can also be set as needed.It is true
It is accurate to protect identification, keyword recognition module carries out in identification process keyword, carries out to all keywords primary again defeated
Enter detection, detects whether as " you are good " and " sorry ", it is first according to Sequential output if front and back detects that state is inconsistent twice
Identification state afterwards.
I.e. state recognition keyword root includes: " you are good ", " sorry ", " number ", " phone ", " extension according to state extraction
Machine ", " dialing ", " network ", " spacing ", " trunk code ", " shutdown ", " in call ", " not yet ", " world ", " leads at " user "
In words ", " being not turned on ", " incoming call limitation ", " arrearage ", " shutdown ", " service ", " ring unanswered ", " shutdown ", between keyword
It is multiplexed and is associated with each other, determine to may recognize that called state through most four times.
In order to form closed loop, keyword recognition module and the state decision discernment module interactive process, the voice is known
Other module carries out keyword recognition according to the keyword that pre-sets, can not be in identified situation in voice, and voice is marked
Know and carry out artificial intervention analysis and be not turned on state, and extracts keyword and be supplemented in state decision discernment module.
Wherein in order to avoid serial keyword repeats, in keyword recognition module detection process, automatic shield is identical
List of keywords, subsequent to again identify that the continuous keyword of the series then automatic shield, guarantee is transferred to state recognition module
Without repetition list of keywords.Such as it is identified as " you are good ", " number ", " spacing " three keywords for the first time, it is subsequent to know again
It is clipped to continuous three keywords then automatic shield, guarantee is transferred to state recognition module without repetition list of keywords.
As shown in Figs 1-4, implementation example is as follows:
1) caller A dials user B by traffic platform.
2) end office (EO) where B starts playback, and the state of B is to converse, and opened calling and kept function, prompts at this time
" you are good, ask should not on-hook, the phone that you dial is busy now ", B has selected on-hook when receiving the traffic request from A
Operation, then prompt becomes " you are good, and the phone that you are dialed is busy now ".
3) traffic platform records to playback.
4) traffic access failure at the end of B normal playback, returns to signaling code, the release recording of traffic platform, and normal recordings do not connect
Logical state;B playback is not finished when not returning to signaling code, and traffic platform stores recording to recording memory module.
5) keyword recognition module shows the spare time to recording memory module, and extracts recorded message.
6) keyword recognition module identidication key, and to keyword message processing, redundancy is eliminated, as repeatedly played
" you are good, ask should not on-hook, the phone that you dial is busy now ", only record primary, recognize that " you are good, the electricity that you are dialed
Words be busy now ", then keyword message record are as follows: " you are good ", on-hook, " phone ", " in call ", " you are good ", " user ",
" just busy ".
7) keyword recognition module by keyword message " you are good ", on-hook, " phone ", " in call ", " you are good ", " use
Family ", " just busy " are sent to state recognition module.
8) state recognition module identifies keyword sequences, can be with identification state are as follows: Called Busy (Call Waiting) ,+
Called Onhook, and return to recorded message and state and give traffic platform.
9) if keyword can not be identified that return label information gives traffic platform by state recognition module.
After failing the information identified by other means analysis such as manual interventions, extracts keyword and improve supplement state recognition
Knowledge base.
Above description is the detailed description for the present invention preferably possible embodiments, but embodiment is not limited to this hair
Bright patent claim, it is all the present invention suggested by technical spirit under completed same changes or modifications change, should all belong to
In the covered the scope of the patents of the present invention.
Claims (8)
1. a kind of communication session call failure scene analysis system, it is characterised in that: traffic platform including safety call center,
Audio storage module, keyword recognition module and state decision discernment module;
Calling subscriber initiates dial-up to called subscriber by the intermediary number of the traffic platform;
The traffic platform carries out speech admission to calling called subscriber's process, after being determined as call failure, by the speech of failure
It stores to audio storage module;
The keyword recognition module identifies the voice audio of failure, and exports keyword;
After the state decision discernment module is by analyzing the keyword of identification, the called shape that can not call connection of output
State.
2. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: it is exhaled in call
It when crying, after media negotiation, records to called playback, when called state playback normal termination and when returning to signaling code, abandons
Recording records called state;At the end of non-normal playback, store to by audio sound-recording to audio storage module.
3. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: the key
After word identification module obtains the audio sound-recording of audio storage module, identification process is interacted with the state decision discernment module
To recording end of identification.
4. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: the state
Decision discernment module is using determine the tree-like branch detection structure of stake, according to the scene of call failure, i.e., spacing, called subscriber shut down,
Called out of reach, caller defaulting subscriber/partial suspension, user apply for hesitation/other reasons hesitation, halted
Machine, Called Busy, Call Waiting, dial again later, called subscriber be provided with incoming call limitation, international long-distance limitation, plus trunk code,
Other operator's direction line failures, long-distance limitation/illegal caller, called long time no answer state carry out corresponding keyword extraction.
5. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: the state
It includes: " you are good " that identidication key is extracted according to state, " sorry ", " number ", " phone ", on-hook, " user ", " is dialled
Beat ", " network ", " spacing ", " trunk code ", " shutdown ", " in call ", " not yet ", " world ", " in call ", " can not connect
It is logical ", " incoming call limitation ", " arrearage ", " shutdowns ", " service ", " ring unanswered ", " shutdown ", be multiplexed between keyword each other and pass
Connection determines to may recognize that called state through most four times.
6. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: the key
Word identification module and the state decision discernment module interactive process, the speech recognition module is according to the key pre-set
Word carries out keyword recognition, can not be in identified situation in voice, and voice is identified and carries out artificial intervention analysis can not connecing
Logical state, and extract keyword and be supplemented in state decision discernment module.
7. a kind of communication session call failure scene analysis system according to claim 4 or 5, it is characterised in that: it is described
In keyword recognition module detection process, the identical list of keywords of automatic shield is subsequent to again identify that the series is continuously closed
Then automatic shield, guarantee are transferred to state recognition module without repetition list of keywords to key word.
8. a kind of communication session call failure scene analysis system according to claim 1, it is characterised in that: the key
Word identification module carry out once re-entering detection to all keywords, detect whether in identification process to keyword
For " you are good " and " sorry ", if front and back detects that state is inconsistent twice, according to the successive identification state of Sequential output.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811054509.9A CN109151220A (en) | 2018-09-11 | 2018-09-11 | A kind of communication session call failure scene analysis system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811054509.9A CN109151220A (en) | 2018-09-11 | 2018-09-11 | A kind of communication session call failure scene analysis system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109151220A true CN109151220A (en) | 2019-01-04 |
Family
ID=64824558
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811054509.9A Pending CN109151220A (en) | 2018-09-11 | 2018-09-11 | A kind of communication session call failure scene analysis system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109151220A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109658939A (en) * | 2019-01-26 | 2019-04-19 | 北京灵伴即时智能科技有限公司 | A kind of telephonograph access failure reason recognition methods |
CN109697243A (en) * | 2019-02-01 | 2019-04-30 | 网易(杭州)网络有限公司 | Ring-back tone clustering method, device, medium and calculating equipment |
CN110379444A (en) * | 2019-08-30 | 2019-10-25 | 北京太极华保科技股份有限公司 | Detection method and device, the electronic equipment of telephone state are judged by preposition media |
CN110995938A (en) * | 2019-12-13 | 2020-04-10 | 上海优扬新媒信息技术有限公司 | Data processing method and device |
CN111147673A (en) * | 2019-12-20 | 2020-05-12 | 北京淇瑀信息科技有限公司 | Method, device and system for cooperatively judging line state by operator signaling and voice |
RU2730693C1 (en) * | 2020-03-11 | 2020-08-25 | Общество с ограниченной ответственностью "Уралинновация" | Voice robotic question-answer system and method of its automatic interaction with electronic device of user |
CN111916071A (en) * | 2020-06-16 | 2020-11-10 | 深圳追一科技有限公司 | Call calling method, device, computer equipment and storage medium |
CN111916109A (en) * | 2020-08-12 | 2020-11-10 | 北京鸿联九五信息产业有限公司 | Feature-based audio classification method and device and computing equipment |
CN112235467A (en) * | 2020-10-09 | 2021-01-15 | 深圳壹账通智能科技有限公司 | Method and device for distinguishing types of disconnected telephones, electronic equipment and storage medium |
CN112866487A (en) * | 2021-02-26 | 2021-05-28 | 上海米鹊科技有限公司 | Method for judging call state of analog telephone set by using voice recognition technology |
CN113079259A (en) * | 2021-03-31 | 2021-07-06 | 北京智齿博创科技有限公司 | Outbound failure result detection method based on freeswitch and ASR technology |
CN113873077A (en) * | 2020-06-30 | 2021-12-31 | 华为技术有限公司 | Call prompting method, call device, readable storage medium and chip system |
CN114584656A (en) * | 2022-05-06 | 2022-06-03 | 北京沃丰时代数据科技有限公司 | Streaming voice response method and device and voice call robot thereof |
CN115334050A (en) * | 2022-08-11 | 2022-11-11 | 迈普通信技术股份有限公司 | Call processing method, device, session initiation protocol server and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101123648A (en) * | 2006-08-11 | 2008-02-13 | 中国科学院声学研究所 | Self-adapted method in phone voice recognition |
JP2013005205A (en) * | 2011-06-16 | 2013-01-07 | Ntt Docomo Inc | Ill-motivated telephone call prevention device and ill-motivated telephone call prevention system |
CN106550155A (en) * | 2016-11-25 | 2017-03-29 | 上海欣方智能系统有限公司 | Suspicious number is carried out swindling the method and system that sample screens classification and interception |
CN106657682A (en) * | 2015-10-31 | 2017-05-10 | 华为技术有限公司 | User state event detection method and apparatus |
CN107580149A (en) * | 2017-08-28 | 2018-01-12 | 携程旅游网络技术(上海)有限公司 | The recognition methods of outgoing call failure cause, device, electronic equipment, storage medium |
-
2018
- 2018-09-11 CN CN201811054509.9A patent/CN109151220A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101123648A (en) * | 2006-08-11 | 2008-02-13 | 中国科学院声学研究所 | Self-adapted method in phone voice recognition |
JP2013005205A (en) * | 2011-06-16 | 2013-01-07 | Ntt Docomo Inc | Ill-motivated telephone call prevention device and ill-motivated telephone call prevention system |
CN106657682A (en) * | 2015-10-31 | 2017-05-10 | 华为技术有限公司 | User state event detection method and apparatus |
CN106550155A (en) * | 2016-11-25 | 2017-03-29 | 上海欣方智能系统有限公司 | Suspicious number is carried out swindling the method and system that sample screens classification and interception |
CN107580149A (en) * | 2017-08-28 | 2018-01-12 | 携程旅游网络技术(上海)有限公司 | The recognition methods of outgoing call failure cause, device, electronic equipment, storage medium |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109658939A (en) * | 2019-01-26 | 2019-04-19 | 北京灵伴即时智能科技有限公司 | A kind of telephonograph access failure reason recognition methods |
CN109697243A (en) * | 2019-02-01 | 2019-04-30 | 网易(杭州)网络有限公司 | Ring-back tone clustering method, device, medium and calculating equipment |
CN110379444A (en) * | 2019-08-30 | 2019-10-25 | 北京太极华保科技股份有限公司 | Detection method and device, the electronic equipment of telephone state are judged by preposition media |
CN110995938A (en) * | 2019-12-13 | 2020-04-10 | 上海优扬新媒信息技术有限公司 | Data processing method and device |
CN110995938B (en) * | 2019-12-13 | 2022-04-26 | 度小满科技(北京)有限公司 | Data processing method and device |
CN111147673A (en) * | 2019-12-20 | 2020-05-12 | 北京淇瑀信息科技有限公司 | Method, device and system for cooperatively judging line state by operator signaling and voice |
RU2730693C1 (en) * | 2020-03-11 | 2020-08-25 | Общество с ограниченной ответственностью "Уралинновация" | Voice robotic question-answer system and method of its automatic interaction with electronic device of user |
WO2021183013A1 (en) * | 2020-03-11 | 2021-09-16 | Индивидуальный Предприниматель Рыков Алексей Яковлевич | Robotic voice question answering system |
CN111916071A (en) * | 2020-06-16 | 2020-11-10 | 深圳追一科技有限公司 | Call calling method, device, computer equipment and storage medium |
CN113873077A (en) * | 2020-06-30 | 2021-12-31 | 华为技术有限公司 | Call prompting method, call device, readable storage medium and chip system |
CN111916109A (en) * | 2020-08-12 | 2020-11-10 | 北京鸿联九五信息产业有限公司 | Feature-based audio classification method and device and computing equipment |
CN111916109B (en) * | 2020-08-12 | 2024-03-15 | 北京鸿联九五信息产业有限公司 | Audio classification method and device based on characteristics and computing equipment |
CN112235467A (en) * | 2020-10-09 | 2021-01-15 | 深圳壹账通智能科技有限公司 | Method and device for distinguishing types of disconnected telephones, electronic equipment and storage medium |
CN112235467B (en) * | 2020-10-09 | 2022-09-02 | 深圳壹账通智能科技有限公司 | Method and device for distinguishing types of disconnected telephones, electronic equipment and storage medium |
CN112866487A (en) * | 2021-02-26 | 2021-05-28 | 上海米鹊科技有限公司 | Method for judging call state of analog telephone set by using voice recognition technology |
CN113079259A (en) * | 2021-03-31 | 2021-07-06 | 北京智齿博创科技有限公司 | Outbound failure result detection method based on freeswitch and ASR technology |
CN114584656B (en) * | 2022-05-06 | 2022-08-09 | 北京沃丰时代数据科技有限公司 | Streaming voice response method and device and voice call robot thereof |
CN114584656A (en) * | 2022-05-06 | 2022-06-03 | 北京沃丰时代数据科技有限公司 | Streaming voice response method and device and voice call robot thereof |
CN115334050A (en) * | 2022-08-11 | 2022-11-11 | 迈普通信技术股份有限公司 | Call processing method, device, session initiation protocol server and storage medium |
CN115334050B (en) * | 2022-08-11 | 2023-09-15 | 迈普通信技术股份有限公司 | Call processing method, device, session initiation protocol server and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109151220A (en) | A kind of communication session call failure scene analysis system | |
US5883945A (en) | Three-way call detection and response system | |
US7012999B2 (en) | Audio caller identification | |
US10194022B2 (en) | System and method for automatically detecting undesired calls | |
US8391445B2 (en) | Caller identification using voice recognition | |
US5327489A (en) | Method and apparatus for monitoring a network for customer signaling during the term of a call | |
CN104936182B (en) | A kind of method and system of intelligence management and control fraudulent call | |
CN109474757A (en) | A kind of phone secretary cloud service implementation method based on the anti-harassing and wrecking of intelligent answer | |
US20030086541A1 (en) | Call classifier using automatic speech recognition to separately process speech and tones | |
US7190768B2 (en) | Method and apparatus for human-assisted adaptive outbound telephone call classification based on pattern recognition | |
CN104093153A (en) | Method for achieving pseudo number call discrimination and interception based on signalling route analysis and system thereof | |
US7099445B2 (en) | Name-alias based delivery of subscriber services | |
CN104427479A (en) | Method and mobile terminal for prompting call state | |
CN103200469B (en) | PON system access terminal realizes the method for ip voice service call emulation | |
CN109428978B (en) | Computer-implemented method for forwarding calls, predictive dialer system | |
US5519768A (en) | Facsimile detection system and method | |
US7221742B1 (en) | Voicemail box with caller-specific storage folders | |
JPH10503632A (en) | Service node for telephone network | |
CN101835124B (en) | Method and system for automatically diagnosing service faults in telecommunication network | |
CN101924845B (en) | Voiceprint recognition technology-based familiarity telephone system and familiarity telephone communication method | |
CN110581924B (en) | Method and system for prompting risk of fraud | |
CN103596070B (en) | The processing method and processing device of configuration information | |
CN110166637A (en) | A kind of spacing recognition methods and device | |
US20030081756A1 (en) | Multi-detector call classifier | |
US6879682B1 (en) | AIN to ISDN message mapping for intelligent peripheral service enhancement |
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 |
Application publication date: 20190104 |
|
RJ01 | Rejection of invention patent application after publication |