Answering machine detection

• US 9,100,479 B2
• Filed: 03/12/2012
• Issued: 08/04/2015
• Est. Priority Date: 03/10/2011
• Status: Active Grant

• Current Assignee: Genesys Telecommunications Laboratories Incorporated (Genesys Cloud Services Incorporated)
• Original Assignee: Angel.com, Inc. (Genesys Cloud Services Incorporated)
• Inventors: Bouzid, Ahmed Tewfik, Durgani, Prakash Bhagwan, Huang, Lei, Lu, Charles Qichi, Snitkovsky, Vadim, Ma, Rong
• Primary Examiner(s): Nguyen, Khai N
• Assistant Examiner(s): Pope, Kharye
• Application Number:
  US13/417,970
• Publication Number:
  US 20120230483A1
• Time in Patent Office: 1,240 Days
• Field of Search: 379/88, 379/266.07, 379/266.08, 379/201.1, 370/352, 370/353, 370/354, 370/355, 370/356, 370/357
• US Class Current: 1/1
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/2027 Live party detection
• CPC Class Codes: H04M 2242/18 Automated outdialling systems
• CPC Class Codes: H04M 3/42161 via computer interface
• CPC Class Codes: H04M 3/5158 in combination with automated outdialling systems devices for signalling identity of wanted subscriber H04M1/26
• CPC Class Codes: H04M 3/533 Voice mail systems

Abstract

A call is placed to a communications device. A response to the call is received from a call recipient associated with the communications device. The received response is processed based on a set of parameters. Based on processing the received response, a determination is made whether the call recipient is a human or an answering machine. If the call recipient is determined to be a human, the call is handled in a first manner. If the call recipient is determined to be an answering machine, the call is handled in a second manner that is different from the first manner.

Claims

1. A method comprising:

receiving, from a first user, a first customization associating first values with a first subset of parameters, wherein the parameters are configured for determining whether a call recipient is a human or a machine, wherein the parameters are preconfigured with different sets of default values corresponding to different call types and the first customization customizes a set of default values corresponding to a first call type, and wherein the first subset of the parameters with the associated first values are stored in a non-transitory data storage medium along with an association to the first user;

placing a first call to a first communications device, wherein the first call is associated with the first user;

receiving a first response to the first call from a first call recipient associated with the first communications device;

processing the first response based on the first subset of the parameters that are associated with the first values;

based on processing the received first response based on the first subset of the parameters with the associated first values, determining whether the first call recipient is a human or a machine;

handling the first call in a first manner if the first call recipient is determined to be a human, and handling the first call in a second manner that is different from the first manner if the first call recipient is determined to be a machine;

receiving, from a second user, a second customization associating second values with a second subset of the parameters, wherein the second subset of the parameters with the associated second values are stored in the non-transitory data storage medium along with an association to the second user;

placing a second call to a second communications device, wherein the second call is associated with the second user;

receiving a second response to the second call from a second call recipient associated with the second communications device;

processing the second response based on the second subset of the parameters that are associated with the second values;

based on processing the second response based on the second subset of the parameters with the associated second values, determining whether the second call recipient is a human or a machine; and

handling the second call in a third manner if the second call recipient is determined to be a human, and handling the second call in a fourth manner if the second call recipient is determined to be a machine.

2. The method of claim 1, wherein the parameters include a maximum time out parameter that indicates a maximum time allowed for processing the received response before a determination is made that the call recipient is a machine, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on the processing of the received response failing to result in a determination that the call recipient is a human or a machine prior to lapse of a time interval equal to the maximum time allowed.

3. The method of claim 1, wherein the parameters include a touch tone parameter that indicates that detection of a dual-tone multi-frequency (DTMF) tone in the received response results in determining that the call recipient is a machine, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on detection of a DTMF tone in the received response, based on the touch tone parameter being selected.

4. The method of claim 1, wherein the parameters include a noise parameter that indicates that a detection of background noise in the received response within a time window subsequent to call pickup and prior to initial detection of speech results in determining that the call recipient is a human, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a human conditioned on the detection of the background noise within the time window, based on the noise parameter being selected.

5. The method of claim 1, wherein the parameters include a silence parameter that indicates that, subsequent to speech being initially detected in the received response, detection of a silence in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a human, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a human conditioned on detecting the silence in the received response subsequent to initial detection of speech in the received response, based on the silence parameter being selected.

6. The method of claim 1, wherein the parameters include a beep time out parameter that indicates a maximum window of time during which monitoring the received response for an audio beep sound will occur and during which time no message will be sent to the call recipient, the maximum window of time beginning after detecting a last spoken word in the received response and ending a predetermined time later, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on detecting a an audio beep in the received response at a time subsequent to the detected last spoken word but prior to the predetermined time, based on the beep time out parameter being selected.

7. The method of claim 1, wherein the parameters include a word count parameter that indicates that spoken words in the received response will be counted and that, if a number of spoken words exceeds a predetermined threshold, a determination will be made that the call recipient is a machine, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on the number of spoken words exceeding the predetermined threshold, based on the word count parameter being selected.

8. The method of claim 1, wherein the parameters include a no speech parameter that indicates that, subsequent to the call being answered, detection of no speech in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a machine, and

wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on detecting no speech in the received response subsequent to the call being picked up, based on the no speech parameter being selected.

9. The method of claim 1, wherein at least one of the first user or the second user is enabled to customize the first subset of the parameters or the second subset of the parameters respectively using a web-enabled interface.

10. The method of claim 1, wherein the machine includes an answering machine.

11. The method of claim 1, wherein the machine includes voicemail.

12. The method of claim 1, wherein the first customization comprises selecting a first call type associated with the first user from the group consisting of a campaign call type, an autodialer call type and a notification call type, and

wherein:

when the first call type is the campaign call type, the first subset of the parameters corresponds to a general setting having predetermined values as the associated first values,

when the first call type is the autodialer call type, the first call type is associated with an autodialer service operable to dial telephone numbers automatically to communicate between two communications devices, and the first subset comprises a predetermined selection of parameters that is different from parameters selected for other call types, the predetermined selection of parameters included in the first subset being adapted to enable determination whether a call recipient is human or machine in less time compared to other call types, and

when the first call type is the notification call type, the first call type is associated with a notification service operable to deliver a pre-recorded message to call recipients, and the first values associated with the first subset comprise predetermined values that provide a longer wait interval compared to other call types before playing the pre-recorded message when a call recipient is a machine.

13. The method of claim 12, wherein the first call type is the campaign call type and the first subset of parameters indicates:

selection of a touch-tone parameter that is adapted to enable determination that a call recipient is a machine based on detecting a dual tone multi-frequency (DTMF) signal in a received response;

de-selection of a word count parameter, the word count parameter indicating that spoken words in the received response will be counted and that, if a number of spoken words exceeds a predetermined threshold, a determination will be made that the call recipient is a machine; and

selection of a beep timeout parameter by setting its value to a non-zero value to enable quicker determination whether a call recipient is a human or machine compared to the autodialer call type; and

wherein the second customization comprises selecting a second call type associated with the second user as the notification call type, and the second subset of the parameters indicates;

selection of the word count parameter; and

selection of the beep timeout parameter by setting its value to a non-zero value.

14. The method of claim 1, wherein the first subset of the parameters with the associated first values and the second subset of the parameters with the associated second values are concurrently stored in the non-transitory data storage medium.

15. The method of claim 1, wherein the first subset and the second subset are different subsets of the parameters.

16. The method of claim 1, wherein the first subset and the second subset are a same subset of the parameters.

17. A system comprising:

one or more processors; and

non-transitory machine-readable storage media in which are embedded instructions for execution by the one or more processors that are configured to cause the one or more processors to perform operations comprising;

receiving, from a first user, a first customization associating first values with a first subset of parameters, wherein the parameters are configured for determining whether a call recipient is a human or a machine, wherein the parameters are preconfigured with different sets of default values corresponding to different call types and the first customization customizes a set of default values corresponding to a first call type, and wherein the first subset of the parameters with the associated first values are stored in a non-transitory data storage medium along with an association to the first user;

placing a first call to a first communications device, wherein the first call is associated with the first user;

receiving a first response to the first call from a first call recipient associated with the first communications device;

processing the first response based on the first subset of the parameters that are associated with the first values;

based on processing the received first response based on the first subset of the parameters with the associated first values, determining whether the first call recipient is a human or a machine;

handling the first call in a first manner if the first call recipient is determined to be a human, and handling the first call in a second manner that is different from the first manner if the first call recipient is determined to be a machine;

receiving, from a second user, a second customization associating second values with a second subset of the parameters, wherein the second subset of the parameters with the associated second values are stored in the non-transitory data storage medium along with an association to the second user;

placing a second call to a second communications device, wherein the second call is associated with the second user;

receiving a second response to the second call from a second call recipient associated with the second communications device;

processing the second response based on the second subset of the parameters that are associated with the second values;

based on processing the second response based on the second subset of the parameters with the associated second values, determining whether the second call recipient is a human or a machine; and

handling the second call in a third manner if the second call recipient is determined to be a human, and handling the second call in a fourth manner if the second call recipient is determined to be a machine.

18. The system of claim 17, wherein the parameters include a maximum time out parameter that indicates a maximum time allowed for processing the received response before a determination is made that the call recipient is a machine, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a machine conditioned on the processing of the received response failing to result in a determination that the call recipient is a human or a machine prior to lapse of a time interval equal to the maximum time allowed.

19. The system of claim 17, wherein the parameters include a touch tone parameter that indicates that detection of a dual-tone multi-frequency (DTMF) tone in the received response results in determining that the call recipient is a machine, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a machine conditioned on detection of a DTMF tone in the received response, based on the touch tone parameter being selected.

20. The system of claim 17, wherein the parameters include a noise parameter that indicates that a detection of background noise in the received response within a time window subsequent to call pickup and prior to initial detection of speech results in determining that the call recipient is a human, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to perform operations comprising determining that the call recipient is a human conditioned on the detection of the background noise within the time window, based on the noise parameter being selected.

21. The system of claim 17, wherein the parameters include a silence parameter that indicates that, subsequent to speech being initially detected in the received response, detection of a silence in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a human, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a human conditioned on detecting the silence in the received response subsequent to initial detection of speech in the received response, based on the silence parameter being selected.

22. The system of claim 17, wherein the parameters include a beep time out parameter that indicates a maximum window of time during which monitoring the received response for an audio beep sound will occur and during which time no message will be sent to the call recipient, the maximum window of time beginning after detecting a last spoken word in the received response and ending a predetermined time later, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a machine conditioned on detecting an audio beep in the received response at a time subsequent to the detected last spoken word but prior to the predetermined time, based on the beep time out parameter being selected.

23. The system of claim 17, wherein the parameters include a word count parameter that indicates that spoken words in the received response will be counted and that, if a number of spoken words exceeds a predetermined threshold, a determination will be made that the call recipient is a machine, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a machine conditioned on the number of spoken words exceeding the predetermined threshold, based on the word count parameter being selected.

24. The system of claim 17, wherein the parameters include a no speech parameter that indicates that, subsequent to the call being answered, detection of no speech in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a machine, and

wherein the instructions that are configured to cause the one or more processors to determine whether the call recipient is a human or a machine include instructions that are configured to cause the one or more processors to determine that the call recipient is a machine conditioned on detecting no speech in the received response subsequent to the call being picked up, based on the no speech parameter being selected.

25. The system of claim 17, wherein at least one of first user or the second user is enabled to customize the first subset of the parameters or the second subset of the parameters respectively using a web-enabled interface.

26. The system of claim 17, wherein the machine includes an answering machine.

27. The system of claim 17, wherein the machine includes voicemail.

28. A computer program product, embodied in a non-transitory computer-readable medium storing software comprising instructions executable by one or more computers, which, upon such execution, cause the one or more computers to perform operations comprising:

receiving, from a first user, a first customization associating first values with a first subset of parameters, wherein the parameters are configured for determining whether a call recipient is a human or a machine, wherein the parameters are preconfigured with different sets of default values corresponding to different call types and the first customization customizes a set of default values corresponding to a first call type, and wherein the first subset of the parameters with the associated first values are stored in a non-transitory data storage medium along with an association to the first user;

placing a first call to a first communications device, wherein the first call is associated with the first user;

receiving a first response to the first call from a first call recipient associated with the first communications device;

processing the first response based on the first subset of the parameters that are associated with the first values;

based on processing the received first response based on the first subset of the parameters with the associated first values, determining whether the first call recipient is a human or a machine;

handling the first call in a first manner if the first call recipient is determined to be a human, and handling the first call in a second manner that is different from the first manner if the first call recipient is determined to be a machine;

receiving, from a second user, a second customization associating second values with a second subset of the parameters, wherein the second subset of the parameters with the associated second values are stored in the non-transitory data storage medium along with an association to the second user;

placing a second call to a second communications device, wherein the second call is associated with the second user;

receiving a second response to the second call from a second call recipient associated with the second communications device;

processing the second response based on the second subset of the parameters that are associated with the second values;

based on processing the second response based on the second subset of the parameters with the associated second values, determining whether the second call recipient is a human or a machine; and

handling the second call in a third manner if the second call recipient is determined to be a human, and handling the second call in a fourth manner if the second call recipient is determined to be a machine.

29. The computer program product of claim 28, wherein the parameters include a maximum time out parameter that indicates a maximum time allowed for processing the received response before a determination is made that the call recipient is a machine, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a machine conditioned on the processing of the received response failing to result in a determination that the call recipient is a human or a machine prior to lapse of a time interval equal to the maximum time allowed.

30. The computer program product of claim 28, wherein the parameters include a touch tone parameter that indicates that detection of a dual-tone multi-frequency (DTMF) tone in the received response results in determining that the call recipient is a machine, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a machine conditioned on detection of a DTMF tone in the received response, based on the touch tone parameter being selected.

31. The computer program product of claim 28, wherein the parameters include a noise parameter that indicates that a detection of background noise in the received response within a time window subsequent to call pickup and prior to initial detection of speech results in determining that the call recipient is a human, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a human conditioned on the detection of the background noise within the time window, based on the noise parameter being selected.

32. The computer program product of claim 28, wherein the parameters include a silence parameter that indicates that, subsequent to speech being initially detected in the received response, detection of a silence in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a human, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a human conditioned on detecting the silence in the received response subsequent to initial detection of speech in the received response, based on the silence parameter being selected.

33. The computer program product of claim 28, wherein the parameters include a beep time out parameter that indicates a maximum window of time during which monitoring the received response for an audio beep sound will occur and during which time no message will be sent to the call recipient, the maximum window of time beginning after detecting a last spoken word in the received response and ending a predetermined time later, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a machine conditioned on detecting a an audio beep in the received response at a time subsequent to the detected last spoken word but prior to the predetermined time, based on the beep time out parameter being selected.

34. The computer program product of claim 28, wherein the parameters include a word count parameter that indicates that spoken words in the received response will be counted and that, if a number of spoken words exceeds a predetermined threshold, a determination will be made that the call recipient is a machine, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a machine conditioned on the number of spoken words exceeding the predetermined threshold, based on the word count parameter being selected.

35. The computer program product of claim 28, wherein the parameters include a no speech parameter that indicates that, subsequent to the call being answered, detection of no speech in the received response having a duration longer than a predetermined threshold results in determining that the call recipient is a machine, and

wherein the instructions that cause the one or more computers to perform operations comprising determining whether the call recipient is a human or a machine comprises instructions that cause the one or more computers to perform operations comprising determining that the call recipient is a machine conditioned on detecting no speech in the received response subsequent to the call being picked up, based on the no speech parameter being selected.

36. The computer program product of claim 28, wherein at least one of the first user or the second user is enabled to customize the first subset of the parameters or the second subset of the parameters respectively using a web-enabled interface.

37. The computer program product of claim 28, wherein the machine includes an answering machine.

38. The computer program product of claim 28, wherein the machine includes voicemail.

39. A method comprising:

receiving, from a first user, a first customization associating values with a first subset of parameters that are configured for determining whether a call recipient is a human or a machine, wherein the first subset includes a touch tone parameter that indicates that detection of a dual-tone multi-frequency (DTMF) tone in a call response results in determining that respective call recipient is a machine, wherein the parameters are preconfigured with different sets of default values corresponding to different call types and the first customization customizes a set of default values corresponding to a first call type, and wherein different users are enabled to customize values for different subsets of the parameters;

placing a call to a communications device, wherein the call is associated with the first user;

receiving a response to the call from a call recipient associated with the communications device;

processing the received response based on the first subset of the parameters, wherein processing the received response based on the first subset of the parameters includes determining that the first user has selected the-touch tone parameter;

upon determining that the first user has selected the touch tone parameter, checking whether the received response includes a DTMF tone;

based on checking whether the received response includes the DTMF tone, determining whether the call recipient is a human or a machine, wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a machine conditioned on detection of the DTMF tone in the received response; and

handling the call in a first manner if the call recipient is determined to be a human, and handling the call in a second manner that is different from the first manner if the call recipient is determined to be a machine.

40. A method comprising:

receiving, from a first user, a first customization associating values with a first subset of parameters that are configured for determining whether a call recipient is a human or a machine, wherein the first subset includes a silence parameter that indicates that, subsequent to speech being initially detected in a received response, detection of a silence in the received response having a duration longer than a predetermined threshold results in determining that respective call recipient is a human, wherein the parameters are preconfigured with different sets of default values corresponding to different call types and the first customization customizes a set of default values corresponding to a first call type, and wherein different users are enabled to customize values for different subsets of the parameters;

placing a call to a communications device, wherein the call is associated with the first user;

receiving a response to the call from a call recipient associated with the communications device;

processing the received response based on the first subset of the parameters, wherein processing the received response based on the first subset of the parameters includes determining that the first user has selected the silence parameter;

upon determining that the first user has selected the silence parameter, checking whether the received response includes silence subsequent to initial detection of speech in the received response;

based on checking whether the received response includes silence subsequent to initial detection of speech in the received response, determining whether the call recipient is a human or a machine, wherein determining whether the call recipient is a human or a machine comprises determining that the call recipient is a human conditioned on detecting the silence in the received response subsequent to initial detection of speech in the received response; and

handling the call in a first manner if the call recipient is determined to be a human, and handling the call in a second manner that is different from the first manner if the call recipient is determined to be a machine.