Method and system for learning call analysis
A system and method are presented for learning call analysis. Audio fingerprinting may be employed to identify audio recordings that answer communications. In one embodiment, the system may generate a fingerprint of a candidate audio stream and compare it against known fingerprints within a database. The system may also search for a speech-like signal to determine if the endpoint contains a known audio recording. If a known audio recording is not encountered, a fingerprint may be computed for the contact and the communication routed to a human for handling. An indication may be made as to if the call is indeed an audio recording. The associated information may be saved and used for future identification purposes.
1. A method for communication learning in a telecommunication system, wherein the telecommunication system comprises at least an automated dialer, a telephony service module, a database, and a media server operatively coupled over a network for exchange of data there between, the method comprising the steps of:
a. selecting, by the automated dialer, a contact from the database, the contact being associated with a telephone number and one or more acoustic fingerprints;
b. retrieving, by the telephony service module, from the database, the one or more acoustic fingerprints and the telephone number associated with the contact;
c. initiating, by the automated dialer, a communication with the contact based on the telephone number, the communication generating audio;
d. analyzing, by the media server, the audio for matches to any of the one or more of the acoustic fingerprints, wherein matches are not identified;
e. routing, via an electronic routing device by the telephony service module, the communication to an agent device associated with an agent for determining whether or not the communication comprises a speech recording;
f. receiving, from the agent device, a signal indicating the communication comprises a speech recording;
g. requesting, by the automated dialer, new acoustic fingerprints from the media server for the speech recording and associating the new acoustic fingerprints with the contact in the database; and
h. disconnecting the communication with the contact after receiving the signal indicating the communication comprises the speech recording.
2. The method of claim 1 , wherein the telephony service module comprises an application programming interface that receives audio recording fingerprints.
3. The method of claim 1 , wherein step (c) further comprises the step of supplying existing fingerprints associated with a contact to a media server.
4. The method of claim 1 , wherein each of the one or more acoustic fingerprints comprises a unique identifier.
5. The method of claim 1 , wherein each of the one or more acoustic fingerprints further comprises information associating the fingerprints with at least one of: voicemail, answering machine, network message, answering service, and audio recording.
6. The method of claim 1 , further wherein step (d) further comprises determining that the audio is from a human.
7. The method of claim 1 , wherein step (e) further comprises the agent associating a disposition of the communication with the contact in the database.
8. The method of claim 7 , wherein the disposition comprises a wrap-up code identifying the communication as comprising a speech recording.
9. A method for communication learning in a telecommunication system, wherein the telecommunication system comprises at least an automated dialer, a telephony service module, a database, and a media server operatively coupled over a network for exchange of data there between, the method comprising the steps of:
a. selecting, by the automated dialer, one or more contacts from the database;
b. performing a lookup, by the automated dialer, for existing acoustic fingerprints associated with each of the one or more contacts in the database;
c. initiating, by the automated dialer, a communication with one of the one or more contacts, the communication generating audio;
d. determining, by the media server, whether any of the existing acoustic fingerprints associated with a contact is present in the audio, wherein matches are not identified;
e. determining, by the media server, a new acoustic fingerprint for the contact;
f. updating, by the automated dialer, a record for the contact in the database by associating the new acoustic fingerprint with the contact in the database;
g. establishing, by an electronic routing device, a connection between a device associated with the contact and a device operated by an agent;
h. receiving, from the device operated by the agent, a signal indicating the audio comprises a speech recording; and
i. disconnecting the device operated by the agent from the device associated with the contact after receiving the signal indicating the audio comprises the speech recording.
10. The method of claim 9 , wherein the telephony service module comprises an application programming interface that receives audio recording fingerprints.
11. The method of claim 9 , wherein step (c) further comprises the step of supplying existing fingerprints associated with a contact to a media server.
12. The method of claim 9 , wherein each of the existing acoustic fingerprints comprises a unique identifier.
13. The method of claim 9 , wherein each existing acoustic fingerprints further comprises information associating the fingerprints with at least one of: voicemail, answering machine, network message, answering service, and audio recording.
14. The method of claim 9 , further wherein step (d) further comprises determining that the audio is from a human.
15. The method of claim 9 , wherein the determining of step (e) further comprises determining the communication comprises a known audio recording which is not associated with the contact.