Answering machine detection (“AMD”) for a contact center by using AMD meta-data
Answering machine detection (“AMD”) processes in a contact center are improved by obtaining and storing AMD meta-data about a known greeting from an automatic voice messaging capability (“AVMC”) on a telephone call to a known number. The AMD meta-data is used in subsequent calls to that known number and the greeting detected is analyzed using the AMD meta-data to make a comparison determination if AMD meta-data obtained from the current greeting matches that stored so as to determine whether the current greeting originated from an AVMC or from a live person. In certain embodiments, a real-time speech analytics (“RTSA”) system is used for processing the greeting to obtain the AMD meta-data which is stored and used when comparing subsequently obtained AMD meta-data. Calls to telephone numbers for which there is no stored AMD meta-data results in analyzing the greeting in order to obtain and store AMD meta-data.
1. A method for determining whether a greeting detected on an answered outbound call from a contact center was provided by an automatic voice messaging capability (AVMC) or a live person, the method comprising:
originating the call from a call handler in the contact center using a call record from a calling list indicating the telephone number, wherein the call record further indicates stored answering machine detection (“AMD”) meta-data is associated with the telephone number; and
processing the greeting by the contact center using the stored AMD meta-data to ascertain the greeting was provided by the AVMC or the live person.
2. The method of claim 1 , wherein the stored AMD meta-data is obtained by processing a prior call to the telephone number, wherein a prior greeting is detected and analyzed to generate the stored AMD meta data.
3. The method of claim 1 , wherein processing the greeting comprises:
determining current AMD meta-data by processing the greeting;
comparing the current AMD meta-data with the stored AMD meta-data; and
ascertaining, based on the comparison of the current AMD meta-data with the stored AMD meta-data, that the greeting was provided by the AVMC.
4. The method of claim 1 , wherein the current AMD meta-data comprises a set of time-based characteristics of the greeting.
5. The method of claim 2 , wherein the stored AMD meta-data is obtained in response to the prior call to the telephone number dispositioned by an agent as having incorrectly determined the prior greeting was provided by a live person.
6. The method of claim 1 , further comprising:
terminating the call to the telephone number in response to ascertaining the greeting was provided by the AVMC, whereby an agent in the contact center is not connected to the call.
7. The method of claim 2 , wherein the processing of the greeting to produce the current AMD meta-data is performed by a real-time speech analytics system bridged onto the call by a call handler in the contact center.
8. A system for determining whether a greeting detected on an answered outbound call from a contact center was provided by an automatic voice messaging capability (AVMC) comprising:
a dialer configured to:
originate the outbound call to a telephone number indicated in a call record, wherein stored Answering Machine Detection (“AMD”) meta-data is associated with the telephone number, and
establish a call leg to a real-time speech analytics (“RTSA”) component, wherein the call leg is bridged onto the call, wherein the greeting comprising audio data is provided to the RTSA on the call leg; and
the RTSA component configured to:
analyze the greeting to determine current AMD meta-data,
compare the current AMD meta-data with the stored AMD meta-data, and
based on the comparison of the current AMD meta-data with the stored AMD meta-data, provide a greeting type indication to the dialer that indicates whether the greeting was provided by the AVMC or a live person answering the call.
9. The system of claim 8 , wherein the dialer is further configured to:
receive the greeting type indication;
if the greeting type indication indicates the greeting was provided by the AVMC, then terminate the call; and
if the greeting type indication indicates the greeting was provided by a live person answering the call, then connect an agent to the call.
10. The system of claim 8 , wherein the RTSA component is further configure to:
generate the stored AMD meta-data by processing a prior greeting received on a prior call to the telephone number.
11. The system of claim 8 , wherein the current AMD meta-data comprises a set of time-based characteristics derived from processing the audio data of the greeting.
12. The system of claim 10 , wherein the RTSA component is further configured to:
generate the stored AMD meta-data in response to the prior call dispositioned by an agent indicating the prior greeting was incorrectly determined to be have been provided by the live person.
13. The system of claim 9 , wherein the dialer is further configured to:
receive a disposition code from the agent after connecting the agent to the call in response to the greeting type indication indicating the greeting was provided by the live person, wherein the disposition code reflects the greeting was incorrectly classified as provided by the live person answering the call; and
instruct the RTSA component to process the greeting to provide updated stored AMD meta-data.
14. The system of claim 13 , wherein the RTSA is further configured to:
replace the stored AMD meta-data with the updated stored AMD meta-data.
15. A non-transitory computer readable medium storing instructions that when executed by a processor cause the processor to:
receive a greeting comprising audio data transmitted on a call leg from a call originated by a dialer to a telephone number, wherein the call has been answered;
cause the greeting to be processed so as to generate current Answering Machine Detection (“AMD”) meta-data representative of the greeting;
receive stored AMD meta-data of a prior call to the telephone number whereby a prior greeting was processed to generate the stored AMD meta-data;
compare the stored AMD meta-data with the current AMD meta-data; and
generate a greeting type indication indicating the greeting was provided by a live person or an automatic voice messaging capability (“AVMC”).
16. The non-transitory computer readable medium of claim 15 wherein causing the greeting to be processed so as to generate the current AMD meta-data further causes the processor to:
generate a set of sequential time-based characteristics of the greeting.
17. The non-transitory computer readable medium of claim 15 , wherein comparing the stored AMD meta-data with the generated AMD meta-data comprises:
determining a difference between a first set of sequential time-based characteristics with a second set of sequential time-based characteristics;
determining the difference is below a threshold; and
generating the greeting type indication indicating the greeting was provided by the AVMC.
18. The non-transitory computer readable medium of claim 15 , wherein comparing the stored AMD meta-data with the generated AMD meta-data comprises:
determining a difference between a first set of sequential time-based characteristics with a second set of sequential time based characteristics;
determining the difference is above a threshold; and
generating the greeting type indication indicating the greeting was provided by the live person.
19. The non-transitory computer readable medium of claim 18 , wherein the instructions further cause the processor to:
cause a second call leg to an agent to be connected to the call in response to the greeting type indication indicating the greeting was provided by the live person.
20. The non-transitory computer readable medium of claim 17 , wherein the instructions further cause the processor to:
terminate the call in response to the greeting type indication indicating the greeting was provided by the AVMC.