US20140334612A1 - Live person detection in an automated calling system - Google Patents

Live person detection in an automated calling system Download PDF

Info

Publication number
US20140334612A1
US20140334612A1 US14/273,340 US201414273340A US2014334612A1 US 20140334612 A1 US20140334612 A1 US 20140334612A1 US 201414273340 A US201414273340 A US 201414273340A US 2014334612 A1 US2014334612 A1 US 2014334612A1
Authority
US
United States
Prior art keywords
call
voice
signal
live
received 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.)
Abandoned
Application number
US14/273,340
Other languages
English (en)
Inventor
Danny Singer
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Noetica Ltd
Original Assignee
Noetica Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Noetica Ltd filed Critical Noetica Ltd
Publication of US20140334612A1 publication Critical patent/US20140334612A1/en
Assigned to NOETICA LTD. reassignment NOETICA LTD. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: SINGER, DANNY
Priority to US15/495,662 priority Critical patent/US10277740B2/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/5158—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with automated outdialling systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/5166—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with interactive voice response systems or voice portals, e.g. as front-ends
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/5183—Call or contact centers with computer-telephony arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2203/00—Aspects of automatic or semi-automatic exchanges
    • H04M2203/20—Aspects of automatic or semi-automatic exchanges related to features of supplementary services
    • H04M2203/2027—Live party detection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2203/00—Aspects of automatic or semi-automatic exchanges
    • H04M2203/30—Aspects of automatic or semi-automatic exchanges related to audio recordings in general
    • H04M2203/306—Prerecordings to be used during a voice call

Definitions

  • This invention relates to automated calling systems for placing telephone calls.
  • the present invention provides a telephony system and a method of operating such a system.
  • FIG. 1 is an overview of a call center system
  • FIG. 2 illustrates the functional software and hardware modules at each agent workstation incorporated in the call center system of FIG. 1 ;
  • FIG. 3 illustrates the functional components of the predictive dialer incorporated in the call center system shown in FIG. 1 together with the message flow between each module;
  • FIGS. 4A to 4D illustrates schematically an algorithm for determining whether an answered call is a live call or whether the response is from an answering machine
  • FIGS. 5A to 5E illustrates schematically the timing of received voice signals and transmitted messages during an answered call as managed by the algorithm of FIG. 4 ;
  • FIG. 6 is an overview of a call center system according to an alternative embodiment of the present invention.
  • the invention has particular application to call center dialing management systems where agent workstations are manned by call agents.
  • Automatic dialing systems place calls to a number of recipients and calls when answered are then connected to agent workstations linked by a Computer Telephony Integration (CTI) system which integrates computer and telephony systems.
  • CTI systems may include features such as automatic call distribution for management of incoming calls and predictive dialing to enable the system to deliver calls which have been answered to an agent as he or she becomes available to take the next call. Predictive dialing thereby enables an agent who has finished dealing with a previous call to be immediately able to pick up a new answered call.
  • EP 03809779 for example describes a predictive dialing system which monitors progress of an agent script which is followed by the agent during a call and which system predicts using statistical analysis the optimum number of outgoing calls to be dialed in order to match the availability of agents.
  • a particular problem with automatic dialing systems is that calls answered automatically by voicemail or answer machines can cause inefficient use of agent's time since the agent may need to spend time listening to the recorded message and manually classifying the call as unproductive.
  • GB 2293723A describes a digital signal processing card for determining the presence of an answering machine. Received audio voice energy is analyzed to detect time periods during which there is a pause in the voice energy and thresholds are applied to the detected time periods to classify the pattern of voice energy as being consistent with either a live call or a response from an answering machine. Calls which are determined to be answered by the recorded voice of an answering machine are accordingly dropped and are not routed to agent terminals whereas calls determined to be live calls are routed to agents for subsequent handling of the call.
  • a disadvantage of this arrangement is that it may take several seconds for the determination to be made, during which time, if the call is a live call, the called party will have listened to a period of silence of several seconds and may hang up with the perception that this is a nuisance call or an automated call for which no agent was available.
  • FIG. 1 illustrates an overview of the operation of a call center system.
  • the call center system includes a number of agent workstations 1 - 1 , 1 - 2 , . . . , 1 - n .
  • Each workstation 1 - 1 , . . . , 1 - n has a respective agent telephone system 2 - 1 , 2 - 2 , . . . , 2 - n and is connected a server computer 4 through a local area network (LAN) 3 .
  • the server computer 4 is connected to a telephony card 12 , which is connected to each of the agent telephone systems 2 - 1 , 2 - 2 , . . .
  • the server computer 4 executes a number of applications including a predictive dialer 11 , a switch 5 having functionality of a PBX (private branch exchange), and a voice recorder 14 .
  • the server computer 4 also has a database 10 storing data including telephone numbers of call destination terminals to be called and also data acquired from completed calls.
  • the predictive dialer 11 is arranged to cause the switch 5 to speed dial the numbers of a number of the call destination terminals 9 - 1 . . . 9 - k connected to the network 7 .
  • the number of calls dialed at any time is dependent on a predicted number of agents who will be available to deal with each call when it is answered by a respondent at a call destination terminal.
  • a telephony card 12 operating as an interface between the server computer 4 and both the carrier network 7 and the agent telephone system 2 - 1 , . . . 2 - n .
  • the switch 5 executed by the server computer 4 allows the server to function as a PBX in routing calls between agent telephone call and destination terminals.
  • the switch 5 has as one of its software components a live call detector 13 which interacts with the telephony card 12 to determine, based on a received signal from a call destination terminal when a telephone call is in progress, whether the received signal represents a live person response in which voice signals originate from detection in real time of a human voice at the call destination terminal.
  • the operation of the live person detector 13 will be described in detail below.
  • a further software component executed by the server computer 4 is the voice recorder 14 which functions to record onto memory a number of messages to be played back under the control of the live call detector 13 as part of a live call detection method.
  • Each agent records a first message R 1 which is a prompt, for instance “hello”;
  • a second message R 2 which is an introduction, for instance “Hi. My name is John and I am calling on behalf of ABC Company to discuss your recent enquiry,” and
  • R 1 , R 2 , R 3 there will be a respective recording for each agent A, B, C, . . . , denoted herein as messages R 1 (A), R 1 (B), R 1 (C); R 2 (A), R 2 (B), R 2 (C), etc.
  • FIG. 2 illustrates the components of each agent workstation 1 .
  • Each workstation comprises a personal computer 21 , a display 25 , a keyboard 27 and a mouse 29 .
  • the personal computer 21 includes a display control unit 33 effective to control the display 25 and a data input unit 35 responsive to signals produced by the user using the keyboard 27 and mouse 29 .
  • Each agent telephone system 2 which is located at the workstation comprises a microphone and headset.
  • the personal computer 21 also includes an agent scripting module 37 including a store 39 storing a series of scripts which may be displayed on the display 25 .
  • Each script is designed to guide the conversation between the call center agent and the telephone respondent for a particular purpose, for example an opinion poll of voting intentions for a forthcoming election, to enable the agent to extract specific information from the respondent. This information may be input into the data input unit 35 via the keyboard 27 and/or mouse during the telephone conversation between the agent and the respondent as will be described in more detail hereafter.
  • Signals indicating that the agent has actually finished the call and has reached an “idle” state are also passed to the predictive dialer 11 .
  • signals indicating that the agent has logged “onto” or “out from” the particular workstation 1 and thus has become available or is no longer available are generated at the relevant times by the agent scripting module 37 and passed through the interface 43 to the predictive dialer 11 .
  • FIG. 3 is an overview of the various functional modules of the predictive dialer 11 .
  • a predictive engine module 51 This module also being effective to send signals back to the agent scripting module 37 in each workstation 1 , these signals being effective to cause a new script stored in the script store 39 to start to run at the workstation 1 .
  • the predictive engine module 51 is also linked to a list management module 53 , a dialing module 55 and a monitor and control module 57 .
  • the list management module 53 stores the numbers to be dialed.
  • the dialing module 55 is effective to send dialing requests to the call center switch 5 and to receive information back from the call center switch 5 as to whether each telephone 9 has been answered.
  • the monitor and control module 57 is effective to provide dialing numbers to the list management module 53 and to provide control parameters to the predictive engine module 51 .
  • the predictive engine module 51 In response to the status signals and log in/log out signals received from the agent workstations and the parameters received from the monitor and control module 57 , the predictive engine module 51 instructs dialing module 55 to dial numbers produced from numbers stored in the list management module 53 .
  • the predictive engine module 51 includes respective interfaces to the agent workstations 1 connected to the LAN 3 , the dialing module 55 , the list management module 53 and the monitor and control module 57 .
  • the predictive engine module 51 also includes an agent status table 69 including data derived from the agent login and/or logout information signals and the “idle” state signals provided from the agent workstations 1 .
  • a statistics unit 71 is provided for maintaining statistics over a predefined period of time, for example one hour.
  • the predictive engine module 51 also includes an algorithm module 73 programmed with an algorithm and the current variable values for use in the algorithm to enable the calculation of the number of respondent phones to be dialed at any particular time from the agent status information stored in the agent status table 69 and the statistical information provided to the predictive engine module from the monitor and control module 57 at that time.
  • algorithm module 73 programmed with an algorithm and the current variable values for use in the algorithm to enable the calculation of the number of respondent phones to be dialed at any particular time from the agent status information stored in the agent status table 69 and the statistical information provided to the predictive engine module from the monitor and control module 57 at that time.
  • the list management module 53 contains respective interfaces to the monitor and control module 57 and the predictive engine module 51 .
  • the list management module 53 includes a relational database system including a list of all of the numbers to be dialed using any particular script stored in the agent workstations 1 , together with a cache storing the dialing numbers to be included in the next list of new numbers to be dialed.
  • the monitor and control module 57 is a user interface and thus includes a display together with a display control unit, a keyboard and mouse together with a data input control system. Interfaces are provided to respectively the list management module 53 and the predictive engine module 51 .
  • the dialing module 55 includes respective interfaces to the predictive engine module 51 and the switch 5 , together with a buffer for storing the next numbers to be dialed.
  • the live call detector 13 is used in conjunction with the telephony card 12 to emulate conversation between an agent and a respondent at a call destination terminal 9 - 1 . . . 9 - n using pre-recorded messages of the voice of the agent, and monitoring the timing of received voice signals relative to the messages. Since the pattern of speech and silence by either party in a typical telephone conversation follows a predictable pattern, which pattern may be regarded as an etiquette of telephone conversation, any departure from the expected pattern can be identified as being consistent with the call being answered by an answering machine.
  • an answering machine will deliver a pattern of speech and silence which ignores any incoming messages and is followed by a continuous tone which is an indicator that the calling party is expected to leave a message after the tone. Detection of such a tone in signals received from the call destination terminal is therefore generally conclusive that an answering machine has responded to the call.
  • any speech from the respondent would not be expected to interrupt speech from the agent. If therefore the system delivers a recorded introductory message which sounds like a live person speaking and it is not interrupted by speech from the respondent, this is taken to be an indication that it is a live call in which a human respondent believes they are in conversation with the agent.
  • a clarification message pre-recorded by the agent is then delivered and, if after the end of the clarification message the respondent speaks within a predetermined waiting period, this taken to be consistent with normal conversation, leading to a determination of a live call in which the human respondent is engaged in conversation.
  • the agent whose voice is being used in the pre-recorded messages is then connected to continue the call.
  • the respondent is therefore unlikely to be aware that the initial phase of providing pre-recorded messages is anything other than a conversation with a live agent.
  • the live call detector 13 is a computer program which in one embodiment is executed in the server computer 4 and which interacts with the telephony card 12 over a network connection 15 using function calls provided by an application program interface (API) in the server computer.
  • API application program interface
  • the telephony card 12 is a proprietary extension card which is user configurable to make use of on-board digital signal processors (DSPs) for signal analysis and signal generation.
  • the telephony card 12 is connected to the server computer 4 by a PCI/cPCI connection and communicates using TCP/IP protocol
  • a telephony card configuration tool is provided in the server computer 4 for the purpose of configuring the function of the telephony card 12 so that an algorithm controlled by the live person detector 13 invokes processes in the telephony card 12 and receives via return function calls from the telephony card 12 the results of signal monitoring and analysis as a telephone call progresses.
  • the algorithm controlled by the live person detector 13 and executed partly in the server computer 4 and partly in the telephony card 12 will now be described with reference to FIGS. 4 and 5 .
  • the predictive dialer 11 initiates the setting up of telephone calls to telephone numbers extracted from the database 10 and telephone calls are placed via the telephony card 12 over the carrier network 7 .
  • the predictive dialer 11 refers to the database 10 to determine whether a live call detection flag has been set for the telephone number to be dialed and only invokes the live call detector 13 if the flag has been set.
  • an answered call is identified at step 401 by the telephony card sensing signaling information in this example from an ISDN line.
  • the telephony card 12 makes a function call to the switch 5 to signal this event and the switch in turn makes a remote procedure call into the predictive dialer 11 so that the predictive dialer then knows that the call has been answered.
  • the predictive dialer 11 checks the status of the agent workstations 1 - 1 to 1 - n to see if any agent is currently waiting for a call dialog to commence. If there is such an agent, the predictive dialer 11 at step 404 assigns to the call the waiting agent. If no agent is currently in a waiting state, step 403 determines which agent is most likely to become available and assigns this call to that agent.
  • the algorithm determines whether, within a period P 1 from the call being answered, any voice signal is detected in the received signal RX from the call destination terminal 9 - 1 .
  • the incoming received signal is digitized and analyzed by the telephony card 12 to categorize the received signal as having one of the following three states;
  • the measured signal amplitude is less than a predetermined threshold and is determined to represent line noise or background noise detected microphonically or otherwise at the call destination terminal 9 - 1 ;
  • tone a continuous sinusoidal tone of a single frequency is detected, the frequency corresponding to one of a set of tones which are characteristic of answering machine tones, the signal being greater than the predetermined threshold and persisting for a minimum of 64 milliseconds.
  • Tone detection may for example be configured to detect frequencies of 1000 Hz and 1,400 Hz which are common frequencies used by answering machines;
  • voice signal the signal waveform has an amplitude greater than the predetermined threshold and does not correspond to one of the above single frequency tones and is therefore classified as corresponding to a spoken word, grunt, or other vocally generated sound, whether live or pre-recorded.
  • the received signal is represented by RX and the transmitted signal by TX, the signal state being represented by position in a horizontal time line progressing from left to right transitioning between a lower level representing silence and an upper level representing the presence of voice signal.
  • the period P 1 referred to above with reference to step 405 is therefore an initial listening period and is configured to have a duration in the range 250 to 1000 milliseconds. If at the end of period P 1 the live call detector 13 determines that no voice signal has been detected, i.e. as monitored by the telephony card 12 the state of the received signal has not transitioned from silence to voice signal, the live person detector 13 responds by retrieving from the voice recorder 14 a recording of a prompt message made by the particular agent assigned to this call, in this case agent A, so that prompt message R 1 (A) in the form of a way file is transferred to the telephony card. The received way file is processed by the telephony card 12 to produce an audio signal transmitted over the telephone line at step 406 in transmitted signal TX to the call destination terminal 9 - 1 .
  • the prompt message R 1 (A) delivered to the call destination terminal 9 - 1 is therefore “hello” which is intended to prompt the recipient, if indeed it is a human recipient, to speak.
  • step 407 follows with a further listening period of duration P 1 , during which any vocal response made by the recipient will be detected in the telephony card 13 by sensing a transition from the silence state to the voice signal state and upon such detection being signaled to the live call detector 13 the algorithm moves forward to a further phase of monitoring commencing at step 410 in FIG. 4B as described below.
  • the next phase of monitoring commences at step 410 in FIG. 48 .
  • the state of the received signal RX is monitored in the telephony card 12 to detect a return transition to the silence state which, upon being communicated by the telephony card 12 to the live call detector 13 , initiates the timing of a further listening period of duration P 2 which in the present embodiment is configured to be in the range 500 to 1500 milliseconds.
  • the live call detector 13 communicates with the predictive dialer 11 , indicating that potentially a live call is in progress, the predictive dialer checking at step 411 whether the currently assigned agent A is busy on another call. If the answer is yes, at step 412 the predictive dialer 11 checks to see if there is an agent in a waiting or assigned state and, if so, selects at step 413 a new agent for the present call.
  • the introduction message R 2 (A) (or R 2 (B) if the new agent is assigned) is transmitted at step 414 to the call destination terminal 9 - 1 , the process of generating the message being similar to that described above in which the way file is obtained by the live call detector 13 and transmitted to the telephony card 12 for reproduction as an analogue audio signal to be output via the ISDN line to the call destination terminal.
  • the introduction message is “Hi, my name is John and I am calling on behalf of ABC Company to discuss your recent enquiry”.
  • the telephony card 12 listens at step 415 for any change in the status of the received signal RX, i.e. listens for a transition to voice signal state which would indicate that the introduction message R 2 (A) was being interrupted by voice signal from the call destination terminal 9 - 1 . Any such interruption could be consistent with the call having been answered by an answering machine, or could be consistent with an accidental interruption by a live person. If no such interruption is detected, the call is determined to be a live call at step 4151 and, provided the assigned agent is determined to be still available at step 416 , the call is transferred at step 418 to the currently assigned agent in response of a signal provided by the live call detector 13 to the predictive dialer 11 .
  • the live call detector 13 at step 4152 signals to the predictive dialer 11 that this call is possibly a live call requiring clarification and at step 417 listens for a further period of silence P 2 before generating a clarification message R 3 (A).
  • the clarification message is generated by the live call detector 13 retrieving the appropriate way file and transmitting the file to the telephony card 12 where it is used to generate at step 419 a signal in the transmitted signal TX.
  • the live call detector determines the call to be a live call at step 4201 since this corresponds to a pattern of speech and silence which is consistent with normal telephone conversation.
  • the clarification gap P 3 is a listening period with a duration set typically in the range 500 to 1,500 milliseconds.
  • the call is then transferred to the predictive dialer 11 to the currently assigned agent at step 424 by action of the predictive dialer 11 , provided that the agent is determined at step 423 not to be already on another call.
  • the recorded clarification message R 3 (A) is again played at step 419 and the step 420 of listening for period P 3 is repeated.
  • This process continues for a number of attempts which, if the number is determined to exceed a limit of L 2 at step 421 , results in the live call detector 13 determining the call to have been answered by an answering machine.
  • L 2 is 2. This is signaled to the predictive dialer 11 which then follows at step 422 a predetermined procedure to abandon the call as being answered by an answering machine.
  • the predictive dialer 11 determines at step 425 whether there is another agent available. If there is more than one agent available, the predictive dialer 11 at step 427 selects the agent according to a preference for an agent who is ready and waiting to one who is only assigned to the call but still dealing with a previous call.
  • the call is then transferred at step 428 by the predictive dialer 11 to the selected agent as a live call. If no agents are available, the call is abandoned at step 426 in accordance with a predetermined procedure by the predictive dialer 11 .
  • the predictive dialer 11 may have a number of alternative strategies for dealing with abandoned calls.
  • a pre-recorded explanatory message is played with an explanation of why the call was made and the lack of available agents, optionally providing a telephone number for the recipient to contact the company to speak to an operative.
  • the live call detector 13 When an answered call is identified at step 401 , the live call detector 13 also commands at step 430 the telephony card 12 to listen for a transition to the tone state in the received signal RX. This activity proceeds in parallel with the above described steps 402 to 428 . If at any time during the call, the transition to tone state is detected at step 431 , this is communicated from the telephony card 12 to the live call detector 13 which in turn signals this event to the predictive dialer 11 . The call is then terminated at step 432 by the predictive dialer 11 .
  • FIG. 5 provides some illustrations in terms of the signal state on received signal RX and transmitted signal TX for a number of cases.
  • FIG. 5A illustrates a case where voice signal 50 is detected shortly after the instant that the progress of the call begins and before the expiry of the initial listening period P 1 .
  • a listening gap P 2 is created after the transition to silence in the received signal RX and then introduction message R 2 (A) recorded by agent A is played in the transmitted signal TX.
  • introduction message R 2 (A) recorded by agent A is played in the transmitted signal TX.
  • the call is determined by the live call detector 13 to be a live call and progresses under the control of the predictive dialer 11 with the agent A engaging in conversation in accordance with a prepared script.
  • FIG. 5B illustrates a second case in which initially there is an absence of voice signal on the received signal RX during the initial listening period P 1 .
  • the prompt message R 1 (A) recorded by agent A is then played and during a subsequent listening period P 1 , a voice signal 51 is detected in the received signal RX.
  • the introduction message R 2 (A) recorded by agent A is then played in the transmitted signal TX. No interruption by way of voice signal on the received signal RX occurs during this playback of message R 2 (A) and the call is therefore determined to be a live call.
  • the progress of the call then continues under the control of the predictive dialer 11 with the agent A engaging in conversation in accordance with a prepared script.
  • FIG. 5C illustrates a third case in which no voice signal is detected in the received signal during an initial listening period P 1 , or during a second period P 1 following a prompt message R 1 (A) recorded by agent A.
  • a voice signal 52 on received signal RX is detected within a further listening period P 1 .
  • the introduction message R 2 (A) recorded by agent A is played and is not interrupted by any voice signal on the received signal RX. Consequently the call is determined to be a live call and continues under the control of the predictive dialer 11 with agent A engaging in conversation with the respondent.
  • FIG. 5D illustrates a fourth case in which voice signal 53 is detected in RX within period P 1 following prompt message R 1 (A) recorded by agent A so that introduction message R 2 (A) recorded by agent A is played in the same way as in the example of Figure B.
  • the respondent interrupts the introductory message as detected by a further voice signal 54 so that, after a listening period P 2 following the end of the voice signal 54 , the clarification message R 3 (A) recorded by agent A is played.
  • a further voice signal 55 is detected in the received signal RX, resulting in the call being determined to be a live call.
  • FIG. 5E illustrates a further case which follows a similar pattern to the example of FIG. 5D except that no voice signal corresponding to voice signal 53 is detected within the period P 3 after the end of the clarification message R 3 (A).
  • the call is determined not to be a live call and to have been answered by an answering machine. The call is then abandoned by the predictive dialer 11 in accordance with a predetermined procedure.
  • the system shown in FIG. 1 including the server computer and telephony card with its associated software is provided as an integral part of the SynthesysTM Voice platform available from Noetica Limited.
  • the telephony card 12 in this example is a Prosody X card available from AculabTM. This system is capable of handling telephone calls to a wide range of telephony devices including IP phones or soft phones using VoIP as well as digital or POTs (plain old telephones) devices in various configurations.
  • Aculab make available libraries of API functions which are utilized in making function calls between the live call detector 13 and telephony card 12 .
  • the software comprising the live call detector 12 is conveniently written in C#(sharp).
  • listener activities which can be set up either to detect continuous tones or voice signals as described above.
  • each of the agents assigned to agent workstations 1 - 1 to 1 - n will at the commencement of their period of duty make a recording of the prompt message R 1 introduction message R 2 and clarification message R 3 .
  • any interaction with a person answering a call will initially include the pre-recorded voice of the agent until such time as the live call detector 13 indicates to the predictive dialer 11 that this is a live call, after which time the agent is connected and the conversation continues according to the script provided to the agent by the agent workstation.
  • the recipient of the call will therefore be unaware that the initial stage of the call is not being handled by a live agent and the transition from recorded voice of the agent to real time voice of the agent should be seamless.
  • the predictive dialer 11 may be configured to modify the script so that the agent provides to the recipient and explanation of the change in voice.
  • the predictive dialer 11 may be configured to only dial a new number when an agent is already in a waiting state, thereby assuring that there is no change in assigned agent between initially dialing the call and determining the call to be a live call.
  • the telephony card 12 may be sold as a hardware product which may be added to existing systems with minimal modification, i.e. simply replacing the existing telephony card with the new card which has the live call detector functionality.
  • FIG. 6 Such an arrangement is shown schematically in FIG. 6 where the live call detector 13 is shown located within the telephony card 12 , the remaining details corresponding generally to those of FIG. 1 .

Landscapes

  • Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Telephonic Communication Services (AREA)
US14/273,340 2013-05-10 2014-05-08 Live person detection in an automated calling system Abandoned US20140334612A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/495,662 US10277740B2 (en) 2013-05-10 2017-04-24 Live person detection in an automated calling system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GB1308508.9A GB2513924A (en) 2013-05-10 2013-05-10 Live person detection in an automated calling system
GB1308508.9 2013-05-10

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/495,662 Continuation US10277740B2 (en) 2013-05-10 2017-04-24 Live person detection in an automated calling system

Publications (1)

Publication Number Publication Date
US20140334612A1 true US20140334612A1 (en) 2014-11-13

Family

ID=48672184

Family Applications (2)

Application Number Title Priority Date Filing Date
US14/273,340 Abandoned US20140334612A1 (en) 2013-05-10 2014-05-08 Live person detection in an automated calling system
US15/495,662 Active 2034-05-21 US10277740B2 (en) 2013-05-10 2017-04-24 Live person detection in an automated calling system

Family Applications After (1)

Application Number Title Priority Date Filing Date
US15/495,662 Active 2034-05-21 US10277740B2 (en) 2013-05-10 2017-04-24 Live person detection in an automated calling system

Country Status (6)

Country Link
US (2) US20140334612A1 (fr)
EP (1) EP2802133B1 (fr)
CA (1) CA2851004C (fr)
ES (1) ES2774554T3 (fr)
GB (1) GB2513924A (fr)
PT (1) PT2802133T (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016092320A1 (fr) * 2014-12-11 2016-06-16 Ultra Communications Ltd Traitement d'appels téléphoniques
US20170230504A1 (en) * 2013-05-10 2017-08-10 Noetica Ltd Live person detection in an automated calling system
US9774736B1 (en) 2016-09-21 2017-09-26 Noble Systems Corporation Augmenting call progress analysis with real-time speech analytics
US9838538B1 (en) * 2016-09-21 2017-12-05 Noble Systems Corporation Using real-time speech analytics to navigate a call that has reached a machine or service
US10250749B1 (en) 2017-11-22 2019-04-02 Repnow Inc. Automated telephone host system interaction
US10887459B1 (en) * 2019-07-17 2021-01-05 Motorola Mobility Llc Identifying a live person on a phone call
US11115528B1 (en) * 2018-01-25 2021-09-07 Amazon Technologies, Inc. Call control service
US11126970B2 (en) * 2017-07-17 2021-09-21 ExpertHiring, LLC Method and system for managing, matching, and sourcing employment candidates in a recruitment campaign

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10014006B1 (en) * 2013-09-10 2018-07-03 Ampersand, Inc. Method of determining whether a phone call is answered by a human or by an automated device
US9667780B1 (en) * 2016-05-05 2017-05-30 Interactive Intelligence Group, Inc. System and method for communication provider and remote device connection analysis
US10277745B1 (en) * 2017-05-30 2019-04-30 Noble Systems Corporation Answering machine detection for a contact center
US10958775B2 (en) 2018-12-10 2021-03-23 Mitel Networks Corporation Speech to dual-tone multifrequency system and method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020051522A1 (en) * 2000-09-01 2002-05-02 Lucas Merrow Speech recognition method of and system for determining the status of an answered telephone during the course of an outbound telephone call
US6850602B1 (en) * 2002-03-27 2005-02-01 Avaya Technology Corp. Method and apparatus for answering machine detection in automatic dialing
US20080015846A1 (en) * 2006-07-12 2008-01-17 Microsoft Corporation Detecting an answering machine using speech recognition
US20080052344A1 (en) * 2006-08-28 2008-02-28 Avaya Technology Llc High availability for voice enabled applications
US20120201362A1 (en) * 2011-02-04 2012-08-09 Google Inc. Posting to social networks by voice
US20130094646A1 (en) * 2007-03-27 2013-04-18 Convergys Cmg Utah, Inc. System and method for the automatic selection of interfaces

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4941168A (en) * 1988-09-21 1990-07-10 U.S. Telecom International Inc. System for the recognition of automated telephone answering devices and delivery of prerecorded messages to such devices
DE19536212B4 (de) 1994-09-28 2004-12-23 Rockwell International Corp., Downers Grove Anordnung zum Erkennen eines Anrufbeantworters
US6021176A (en) * 1996-02-20 2000-02-01 Solopoint, Inc. Method for monitoring incoming calls to a user's premises on a pots telephone line
US6324262B1 (en) * 1998-03-26 2001-11-27 Market Ability, Inc. Method and system for automated delivery of nontruncated messages
US6925166B1 (en) * 2000-01-12 2005-08-02 Avaya Technology Corp. Method and apparatus for generating automatic greetings in a call center
US7054419B2 (en) * 2001-01-02 2006-05-30 Soundbite Communications, Inc. Answering machine detection for voice message delivery method and system
GB2385487B (en) 2002-10-29 2004-01-21 Noetica Ltd Call centre dialling management system
US7184521B2 (en) * 2004-06-10 2007-02-27 Par3 Communications, Inc. Method and system for identifying a party answering a telephone call based on simultaneous activity
GB2478916B (en) 2010-03-22 2014-06-11 Veritape Ltd Transaction security method and system
WO2012122568A1 (fr) * 2011-03-10 2012-09-13 Ahmed Tewfik Bouzid Détection de répondeur téléphonique
GB2513924A (en) * 2013-05-10 2014-11-12 Noetica Ltd Live person detection in an automated calling system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020051522A1 (en) * 2000-09-01 2002-05-02 Lucas Merrow Speech recognition method of and system for determining the status of an answered telephone during the course of an outbound telephone call
US20130108031A1 (en) * 2000-09-01 2013-05-02 Eliza Corporation Speech Recognition Method of and System for Determining the Status of an Answered Telephone During the Course of an Outbound Telephone Call
US6850602B1 (en) * 2002-03-27 2005-02-01 Avaya Technology Corp. Method and apparatus for answering machine detection in automatic dialing
US20080015846A1 (en) * 2006-07-12 2008-01-17 Microsoft Corporation Detecting an answering machine using speech recognition
US20080052344A1 (en) * 2006-08-28 2008-02-28 Avaya Technology Llc High availability for voice enabled applications
US20130094646A1 (en) * 2007-03-27 2013-04-18 Convergys Cmg Utah, Inc. System and method for the automatic selection of interfaces
US20120201362A1 (en) * 2011-02-04 2012-08-09 Google Inc. Posting to social networks by voice

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170230504A1 (en) * 2013-05-10 2017-08-10 Noetica Ltd Live person detection in an automated calling system
US10277740B2 (en) * 2013-05-10 2019-04-30 Noetica Ltd. Live person detection in an automated calling system
WO2016092320A1 (fr) * 2014-12-11 2016-06-16 Ultra Communications Ltd Traitement d'appels téléphoniques
US10084915B1 (en) 2016-09-21 2018-09-25 Noble Systems Corporation Augmenting call progress analysis with real-time speech analytics
US9961201B1 (en) 2016-09-21 2018-05-01 Noble Systems Corporation Using real-time speech analytics to navigate a call that has reached a machine or service
US10044859B1 (en) 2016-09-21 2018-08-07 Noble Systems Corporation Using real-time speech analytics to navigate a call that has reached a machine or service
US9838538B1 (en) * 2016-09-21 2017-12-05 Noble Systems Corporation Using real-time speech analytics to navigate a call that has reached a machine or service
US9774736B1 (en) 2016-09-21 2017-09-26 Noble Systems Corporation Augmenting call progress analysis with real-time speech analytics
US11126970B2 (en) * 2017-07-17 2021-09-21 ExpertHiring, LLC Method and system for managing, matching, and sourcing employment candidates in a recruitment campaign
US10250749B1 (en) 2017-11-22 2019-04-02 Repnow Inc. Automated telephone host system interaction
US10432790B2 (en) 2017-11-22 2019-10-01 Repnow Inc. Automated telephone host system interaction
US10477022B2 (en) 2017-11-22 2019-11-12 Repnow Inc. Automated telephone host system interaction
US11025778B2 (en) 2017-11-22 2021-06-01 Repnow Inc. Automated telephone host system interaction
US11115528B1 (en) * 2018-01-25 2021-09-07 Amazon Technologies, Inc. Call control service
US10887459B1 (en) * 2019-07-17 2021-01-05 Motorola Mobility Llc Identifying a live person on a phone call

Also Published As

Publication number Publication date
ES2774554T3 (es) 2020-07-21
CA2851004A1 (fr) 2014-11-10
EP2802133B1 (fr) 2020-02-05
PT2802133T (pt) 2020-06-08
CA2851004C (fr) 2020-12-22
US10277740B2 (en) 2019-04-30
GB2513924A (en) 2014-11-12
US20170230504A1 (en) 2017-08-10
EP2802133A1 (fr) 2014-11-12
GB201308508D0 (en) 2013-06-19

Similar Documents

Publication Publication Date Title
US10277740B2 (en) Live person detection in an automated calling system
US10110741B1 (en) Determining and denying call completion based on detection of robocall or telemarketing call
US8054964B2 (en) System and method for detecting emotions at different steps in a communication
US6600821B1 (en) System and method for automatically detecting problematic calls
US11240374B2 (en) Call processing method and apparatus, server, storage medium, and system
US20190098135A1 (en) System, Method, and Apparatus for Determining a Status of a Call Recipient in a Call System
US9583108B2 (en) Voice detection for automated communication system
US20050276390A1 (en) Method and system for identifying a party answering a telephone call based on simultaneous activity
US8391320B2 (en) State-based management of messaging system jitter buffers
US20210274044A1 (en) Automated telephone host system interaction
US11516341B2 (en) Telephone call screener based on call characteristics
US20130151248A1 (en) Apparatus, System, and Method For Distinguishing Voice in a Communication Stream
CN114420130A (zh) 电话语音交互方法、装置、设备及存储介质
WO2019103745A1 (fr) Interaction automatisée d'un système hôte téléphonique
US20090245487A1 (en) Device for Alerting a Telephone Caller When a Live Person is Reached
EP1565907A2 (fr) Procedes et apparatus de surveillance et d'evaluation de donnees au moyen d'une reconnaissance vocale
WO2014152542A2 (fr) Détection de voix pour un système de communication automatisé
EP1554719A2 (fr) Procedes et appareil d'analyse de donnees audio et d'exploration de donnees au moyen de la reconnaissance vocale

Legal Events

Date Code Title Description
AS Assignment

Owner name: NOETICA LTD., UNITED KINGDOM

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SINGER, DANNY;REEL/FRAME:034953/0956

Effective date: 20141103

STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION