System and method for three-way call detection
A system for detecting three-way calls in a monitored telephone conversation includes a speech recognition processor that transcribes the monitored telephone conversation and associates characteristics of the monitored telephone conversation with a transcript thereof, a database to store the transcript and the characteristics associated therewith, and a three-way Call detection processor to analyze the characteristics of the conversation and to detect therefrom the addition of one or more parties to the conversation. The system preferably includes at least one domain-specific language model that the speech recognition processor utilizes to transcribe the conversation. The system may operate in real-time or on previously recorded conversations. A query and retrieval system may be used to retrieve and review call records from the database.
1. A system for detecting a three-way call in a monitored telephone conversation, the system comprising:
one or more circuits configured to:
retrieve content of the monitored telephone conversation from a file storage system;
generate a domain specific language model associated with a participant of the monitored telephone conversation;
convert a spoken audio of the content of the monitored telephone conversation to a transcript using the domain specific language model;
detect a first audio characteristic at a first timing and a second audio characteristic at a second timing through an analysis of the content of the monitored telephone conversation, including analysis of at least one of the spoken audio and the transcript of the monitored telephone conversation;
assign a first value of the first audio characteristic based on a predefined importance of the first audio characteristic;
assign a second value of the second audio characteristic based on a predefined importance of the second audio characteristic and a temporal difference between the first timing and the second timing; and
detect the three-way call based on the first value and the second value,
wherein the first audio characteristic and the second audio characteristic are indicators of the three-way call.
2. The system of claim 1 , further comprising:
a database coupled to the one or more circuits, the database being configured to store the first value and the second value.
3. The system of claim 1 , wherein the monitored telephone conversation is one of a monitored voice over IP (VOIP) telephone call or a monitored plain old telephone system (POTS).
4. The system of claim 1 , wherein the one or more circuits are further configured to calculate a three-way calling score by normalizing a sum of the first value and the second value to a first number between 0 and 1 using a refining function, and
wherein the refining function includes:
calculating a second number as the natural log of the negative of the sum;
adding 1 to the second number to obtain a third number; and
inverting the third number.
5. The system of claim 1 , wherein at least one of the first audio characteristic and the second audio characteristic is a verbal characteristic spoken during the monitored telephone conversation.
6. The system of claim 1 , wherein the first audio characteristic represents a detection of a spoken phone number during the monitored telephone conversation, and
wherein the second audio characteristic represents a gap in speech during the monitored telephone conversation.
7. The system of claim 1 , wherein an amount of the second value increases based on a temporal proximity of the second audio characteristic to the first audio characteristic within the monitored telephone conversation.
8. The system of claim 1 , wherein the first audio characteristic includes one or more gaps in the monitored telephone conversation, dial tones, dial pulses, ring tones, or telephone salutations.
9. The system of claim 1 , wherein the second audio characteristic includes one or more of a frequency of phrases uttered during the monitored telephone conversation, a third timing of phrases uttered during the monitored telephone conversation, or a fourth timing of extracted telephone conversation characteristics.
10. The system of claim 1 , wherein the one or more circuits are further configured to generate a transcript of the monitored telephone conversation, and to tag the transcript with a timestamp associated with detection of the three-way call in the monitored telephone conversation.
11. The system of claim 1 , wherein at least one of the first value or the second value is weighted based on the first timing of the first audio characteristic in relation to the second timing of the second audio characteristic.
12. The system of claim 11 , wherein the one or more circuits are further configured to compute a three-way calling score based on the first value and the second value.
13. The system of claim 12 , wherein the one or more circuits are further configured to detect the three-way call based on a comparison of the three-way calling score to a predetermined threshold.
14. The system of claim 1 , wherein the one or more circuits are further configured to detect the three-way call by calculating a likelihood of a presence of a third party using an algorithm based on the first value and the second value.
15. A method for detecting a three-way call in a monitored telephone conversation, the method comprising:
retrieving content of the monitored telephone conversation from a file storage system;
generating a domain specific language model associated with a participant of the monitored telephone conversation;
converting a spoken audio of the content of the monitored telephone conversation to a transcript using the domain specific language model;
detecting, by one or more circuits, a first audio characteristic at a first timing and a second audio characteristic at a second timing through analyzing of the content of the monitored telephone conversation, including analyzing of at least one of the spoken audio and the transcript of the monitored telephone conversation;
assigning a first value of the first audio characteristic based on a predefined importance of the first audio characteristic;
assigning a second value of the second audio characteristic based on a predefined importance of the second audio characteristic and a temporal difference between the first timing and the second timing; and
detecting the three-way call based on the first value and the second value,
wherein the first audio characteristic and the second audio characteristic are indicators of the three-way call.
16. The method of claim 15 , further comprising:
storing the first value and the second value in a database.
17. The method of claim 15 , wherein the monitored telephone conversation is one of a monitored voice over IP (VOIP) telephone call or a monitored plain old telephone system (POTS).
18. The method of claim 15 , further comprising calculating a three-way calling score by normalizing a sum of the first value and the second value to a first number between 0 and 1 by using a refining function,
wherein the using the refining function includes:
calculating a second number as the natural log of the negative of the sum;
adding 1 to the second number to obtain a third number; and
inverting the third number.
19. The method of claim 15 , wherein at least one of the first audio characteristic and the second audio characteristic is a verbal characteristic spoken during the monitored telephone conversation.
20. The method of claim 15 , wherein the first audio characteristic represents a detection of a spoken phone number during the monitored telephone conversation, and
wherein the second audio characteristic represents a gap in speech during the monitored telephone conversation.
21. The method of claim 15 , wherein an amount of the second value increases based on a temporal proximity of the second audio characteristic to the first audio characteristic within the monitored telephone conversation.
22. The method of claim 15 , wherein the first audio characteristic includes one or more gaps in the monitored telephone conversation, dial tones, dial pulses, ring tones, or telephone salutations.
23. The method of claim 15 , wherein the second audio characteristic includes one or more of a frequency of phrases uttered during the monitored telephone conversation, a third timing of phrases uttered during the monitored telephone conversation, or a fourth timing of extracted telephone conversation characteristics.
24. The method of claim 15 , further comprising:
generating a transcript of the monitored telephone conversation; and
tagging the transcript with a timestamp associated with detection of the three-way call in the monitored telephone conversation.
25. The method of claim 15 , wherein at least one of the first value or the second value is weighted based on the first timing of the first audio characteristic in relation to the second timing of the second audio characteristic.
26. The method of claim 25 , further comprising computing a three-way calling score based on the first value and the second value.
27. The method of claim 26 , further comprising detecting the three-way call based on a comparison of the three-way calling score to a predetermined threshold.
28. The method of claim 15 , wherein the detecting the three-way call includes calculating a likelihood of a presence of a third party using an algorithm based on the first value and the second value.
29. The system of claim 2 , wherein the file storage system is a part of the database.
30. The method of claim 16 , wherein the file storage system is a part of the database.