Multi-detector call classifier

• US 20030081756A1
• Filed: 10/23/2001
• Published: 05/01/2003
• Est. Priority Date: 10/23/2001
• Status: Abandoned Application

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Spencer, Douglas A., Chan, Norman C.
• Application Number:
  US10/037,583
• Publication Number:
  US 20030081756A1
• Time in Patent Office: Days
• US Class Current: 379/243
• 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 3/2218 Call detail recording
• CPC Class Codes: H04M 3/487 Arrangements for providing information services, e.g. recorded voice services or time announcements
• CPC Class Codes: H04M 3/5158 in combination with automated outdialling systems devices for signalling identity of wanted subscriber H04M1/26
• CPC Class Codes: H04Q 1/44 using alternate current H04Q1/50 takes precedence

Abstract

Classifying a call to a called destination endpoint by a call classifier. The call classifier is responsive to information received from the called destination endpoint to perform the call classification.

Claims

1. An apparatus for classifying a call to a destination endpoint comprising:

a receiver for receiving information from the destination endpoint;

a first detector for determining a first classification in response to the information received from the destination endpoint;

a second detector for determining a second classification in response to the information received from the destination endpoint;

a third detector for determining a third classification in response to the information received from the destination endpoint; and

an inference engine for determining a call classification of the destination endpoint in response to the first, second, and third classifications.

2. The apparatus of claim 1 further comprises a fourth detector for determining a fourth classification in response to the information received from the destination endpoint;

and

the inference engine further responsive to the fourth classification for determining the call classification of the destination endpoint.

3. The apparatus of claim 1 wherein the first detector is a tone detector.

4. The apparatus of claim 1 wherein the second detector is an energy analyzer.

5. The apparatus of claim 1 wherein the third detector is a zero crossing analyzer.

6. The apparatus of claim 2 wherein the fourth detector is an automatic speech recognizer.

7. The apparatus of claim 6 further comprises a recorder for recording the received information and for updating the inference engine.

8. The apparatus of claim 2 wherein the first detector is a tone detector, the second detector is an energy analyzer, and third detector is a zero crossing analyzer;

9. The apparatus of claim 8 wherein the fourth detector is an automatic speech recognizer.

10. A call classifier for classifying a call to a destination endpoint comprising:

a circuit for receiving information from the destination endpoint and for processing the received information;

a tone detector for determining a first classification in response to the processed information;

a energy analyzer detector for determining a second classification in response to the processed information;

a zero crossing analyzer detector for determining a third classification in response to the processed information; and

an inference engine for determining a call classification of the destination endpoint in response to the first, second, and third classifications.

11. The call classifier of claim 10 further comprises a recorder for recording the received information and for updating the inference engine.

12. A call classifier for classifying a call to a destination endpoint comprising:

a circuit for receiving information from the destination endpoint and for processing the received information;

a tone detector for determining a first classification in response to the processed information;

a energy analyzer detector for determining a second classification in response to the processed information;

a zero crossing analyzer detector for determining a third classification in response to the processed information;

an automatic speech recognition unit for determining a fourth classification; and

an inference engine for determining a call classification of the destination endpoint in response to the first, second, third and fourth classifications.

13. The call classifier of claim 12 further comprises a recorder for recording the received information and for updating the inference engine.

14. The call classifier of claim 12 wherein the automatic speech recognition unit is determining words.

15. The call classifier of claim 12 wherein the automatic speech recognition unit is determining phrases.

16. The call classifier of claim 15 wherein the automatic speech recognition unit is executing a Hidden Markov Model.

17. A method for classifying a call to a destination endpoint, comprising the steps of:

receiving information from the called destination endpoint;

performing a first classification of the received information;

performing a second classification of the received information;

performing a third classification of the received information; and

determining a call classification of the called destination endpoint from the first, second, and third classifications.

18. The method of claim 17 further comprises the step of performing a fourth classification of the received information;

and

the step of determining further responsive to the fourth classification to determine the call classification of the called destination endpoint.

19. The method of claim 18 wherein the first classification is for one of tone, energy, zero crossings, or speech.

20. The method of claim 19 wherein the second classification is for one of tone, energy, zero crossings, or speech.

21. The method of claim 19 wherein the third classification is for one of tone, energy, zero crossings, or speech.

22. The method of claim 21 wherein the fourth classification is for one of tone, energy, zero crossings, or speech.

23. The method of claim 22 wherein the step of determining comprises the step of executing an inference engine.

24. The method of claim 23 further comprises the step of recording the received information for updating the inference engine.

25. The method of claim 23 wherein performing classification for speech comprises the step of executing a Hidden Markov Model.

26. The method of claim 23 wherein performing classification for speech comprises the step of determining words.

27. The method of claim 23 wherein performing classification for speech comprises the step of determining phrases.

32. Apparatus for implementing the steps of claim 17.

33. Apparatus for implementing the steps of claim 18.

28. A method for classifying a call to a destination endpoint, comprising the steps of:

receiving information from the called destination endpoint;

performing a tone classification of the received information;

performing a energy classification of the received information;

performing a zero crossing classification of the received information;

performing speech classification of the received information; and

executing an inference engine to determine a call classification of the called destination endpoint from the tone, energy, zero crossing, and speech classifications.

29. The method of claim 28 wherein performing speech classification comprises the step of determining words.

30. The method of claim 28 wherein performing speech classification comprises the step of determining phrases.

31. The method of claim 28 further comprises the step of recording the received information for updating the inference engine.