Systems and methods for detecting inmate to inmate conference calls

• US 10,129,392 B1
• Filed: 08/25/2017
• Issued: 11/13/2018
• Est. Priority Date: 08/25/2017
• Status: Active Grant

• Current Assignee: Global Tel-Link Corporation
• Original Assignee: Global Tel-Link Corporation
• Inventors: Hodge, Stephen Lee
• Primary Examiner(s): Patel, Hemant
• Application Number:
  US15/686,938
• Time in Patent Office: 445 Days
• Field of Search: 379 671- 8828, 37920101, 37920201-20701, 379 37- 52, 370259-271, 370351-357, 4554121-4261, 704270-278, 709201-207, 709217-248
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 25/27 characterised by the analysis technique
• 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 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 2203/301 Management of recordings
• CPC Class Codes: H04M 2203/556 Statistical analysis and interpretation
• CPC Class Codes: H04M 2203/558 Databases
• CPC Class Codes: H04M 2203/6054 Biometric subscriber identification
• CPC Class Codes: H04M 3/2281 Call monitoring, e.g. for law enforcement purposes; Call tracing; Detection or prevention of malicious calls
• CPC Class Codes: H04M 3/56 Arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities video conference systems H04N7/15

Abstract

A system for detecting inmate to inmate conference calls in a correctional facility is disclosed herein. The system includes a database and a conference call detection server, wherein the conference call detection server is configured to monitor a plurality of inmate communications, convert an audio signal of each inmate communication to a frequency domain signal, identify frequency data comprising one or more frequency peaks and corresponding frequency values in the frequency domain signal for each inmate communication, generate a record comprising the frequency data for each inmate communication, resulting in a plurality of records, store the plurality of records in the database, detect an inmate to inmate conference call by matching a frequency subset of a new inmate communication with frequency data in a detected record in the database, and verify the inmate to inmate conference call by matching audio with voice biometric samples.

Claims

1. A method for detecting inmate to inmate conference calls, the method comprising:

monitoring a plurality of inmate communications by an inmate conference call detection server, wherein each inmate communication comprises an audio signal;

converting the audio signal of each inmate communication to a frequency domain signal over a first predetermined window of time for each inmate communication;

identifying one or more frequency peaks in the frequency domain signal for each inmate communication and a frequency value corresponding to each frequency peak;

generating a record for each inmate communication comprising frequency data regarding the one or more frequency peaks and frequency values for each inmate communication, resulting in a plurality of records for the plurality of inmate communications, wherein the plurality of records comprises at least one record associated with a first inmate;

storing the plurality of records in an inmate communication database;

monitoring a new inmate communication at an inmate communication device of a second inmate, wherein the new inmate communication comprises a new audio signal of the second inmate;

converting the new audio signal to a new frequency domain signal over a second predetermined window of time for the new inmate communication;

determining that a portion of the new frequency domain signal for the new inmate communication matches frequency data in a detected record in the inmate communication database, wherein the detected record is associated with the first inmate and comprises a detected audio signal;

detecting an inmate to inmate conference call between the first inmate and the second inmate based on the determination; and

in response to the detection, verifying the inmate to inmate conference call by matching the detected audio signal with a first voice biometric sample of the first inmate and the new audio signal with a second voice biometric sample of the second inmate.

2. The method of claim 1, further comprising:

converting the audio signal of each inmate communication from a time domain signal to the frequency domain signal by applying a Fourier transform to the time domain signal.

3. The method of claim 1, wherein the identifying one or more frequency peaks in the frequency domain signal for each inmate communication further comprises:

applying a peak detection algorithm to the frequency domain signal for each inmate communication.

4. The method of claim 1, wherein the identifying one or more frequency peaks in the frequency domain signal for each inmate communication further comprises:

identifying a first frequency peak in the frequency domain signal, a first frequency value corresponding to the first frequency peak, and a first amplitude value of the first frequency peak; and

identifying a second frequency peak in the frequency domain signal, a second frequency value corresponding to the second frequency peak, and a second amplitude value of the second frequency peak, wherein the first amplitude value and the second amplitude value are the same.

5. The method of claim 4, further comprising:

storing data regarding the first frequency peak, the first frequency value, the second frequency peak, and the second frequency value for each inmate communication in a first record in the inmate communication database.

6. The method of claim 1, wherein the determining that a portion of the new frequency domain signal for the new inmate communication matches frequency data in a detected record in the inmate communication database further comprises:

matching frequency peaks and frequency values in the portion of the new frequency domain signal with a first subset of frequency data stored in a first detected record;

identifying a number of frequency matches based on the matching; and

determining whether the number of frequency matches is above or below a predetermined threshold value.

7. The method of claim 6, further comprising:

in response to determining that the number of frequency matches is above the predetermined threshold value, detecting the inmate to inmate conference call between the first inmate associated with the first subset of frequency data and the second inmate associated with the new inmate communication.

8. The method of claim 6, further comprising:

in response to determining that the number of frequency matches is below the predetermined threshold value, matching the frequency peaks and frequency values in the portion of the new frequency domain signal with a second subset of frequency data stored in a second detected record.

9. A system for detecting inmate to inmate conference calls in correctional facilities, the system comprising:

an inmate communication database storing a plurality of records corresponding to a plurality of inmate communications at the correctional facilities, wherein the plurality of records includes at least one record associated with a first inmate;

an inmate communication device associated with a second inmate;

a conference call detection server comprising at least one processor and a memory; and

a network that communicatively couples the inmate communication device, the inmate communication database, and the conference call detection server, wherein the at least one processor of the conference call detection server is configured to;

receive an inmate communication of the second inmate from the inmate communication device, wherein the inmate communication comprises an audio signal;

convert the audio signal of the inmate communication to a frequency domain signal over a predetermined window of time;

determine that a portion of the frequency domain signal for the inmate communication matches frequency data in a detected record in the inmate communication database, wherein the detected record is associated with the first inmate and comprises a detected audio signal;

detect an inmate to inmate conference call between the first inmate and a second inmate based on the determination; and

in response to the detection, verify the inmate to inmate conference call by matching the detected audio signal with a first voice biometric sample of the first inmate and the audio signal with a second voice biometric sample of the second inmate.

10. The system of claim 9, wherein each record in the plurality of records comprises frequency data regarding one or more frequency peaks and frequency values for each inmate communication.

11. The system of claim 10, wherein each record in the plurality of records further comprises time data regarding time points at which each frequency peak occurred for each inmate communication.

12. The system of claim 9, wherein the determining that a portion of the frequency domain signal for the inmate communication matches frequency data in a detected record in the inmate communication database further comprises:

matching frequency peaks and frequency values of the portion of the frequency domain signal with a first subset of frequency data stored in a first detected record;

identifying a number of frequency matches based on the matching; and

determining whether the number of frequency matches is above or below a predetermined threshold value.

13. The system of claim 12, wherein the at least one processor of the conference call detection server is further configured to:

detect the inmate to inmate conference call between the first inmate and a second inmate associated with the first subset of frequency data in response to determining that the number of frequency matches is above the predetermined threshold value.

14. The system of claim 12, wherein the at least one processor of the conference call detection server is further configured to:

match the frequency peaks and frequency value in the portion of the frequency domain signal with a second subset of frequency data stored in a second detected record in response to determining that the number of frequency matches is below the predetermined threshold value.

15. A system for identifying frequency data for detection of inmate to inmate conference calls in correctional facilities:

an inmate communication database;

a conference call detection server comprising at least one processor and a memory storing a peak detection module and a voice biometrics module;

a network that communicatively couples the inmate communication database and the conference call detection server,

wherein the at least one processor of the conference call detection server is configured to;

monitor a plurality of inmate communications, wherein each inmate communication comprises an audio signal;

convert the audio signal of each inmate communication to a frequency domain signal over a predetermined window of time for each inmate communication;

identify one or more frequency peaks in the frequency domain signal for each inmate communication and a frequency value corresponding to each frequency peak by applying the peak detection module to the frequency domain signal for each inmate communication;

generate a record for each inmate communication comprising frequency data regarding the one or more frequency peaks and frequency values for each inmate communication and time data regarding time points at which each frequency peak occurred for each inmate communication, resulting in a plurality of records for the plurality of inmate communications;

store the plurality of records in the inmate communication database;

detect an inmate to inmate conference call between a first inmate and a second inmate by matching a frequency subset of a new inmate communication with frequency data in a detected record in the inmate communication database, wherein the detected record comprises a detected audio signal; and

in response to the detection, verify the inmate to inmate conference call by using the voice biometrics module to match the detected audio signal with a first voice biometric sample of the first inmate and an audio sample of the new inmate communication with a second voice biometric sample of the second inmate.

16. The system of claim 15, wherein the new inmate communication comprises a time domain audio signal that is converted to a frequency domain signal comprising the frequency sub set.

17. The system of claim 15, wherein the detecting the inmate to inmate conference call further comprises:

matching frequency peaks and frequency values in the frequency subset of the new inmate communication with a first subset of frequency data stored in a first detected record in the inmate communication database;

identifying a number of frequency matches based on the matching; and

determining whether the number of frequency matches is above or below a predetermined threshold value.

18. The system of claim 17, wherein the at least one processor of the conference call detection server is further configured to:

in response to determining that the number of frequency matches is above the predetermined threshold value, detect the inmate to inmate conference call between the first inmate associated with the first subset of frequency data and the second inmate associated with the new inmate communication.

19. The system of claim 17, wherein the at least one processor of the conference call detection server is further configured to:

in response to determining that the number of frequency matches is below the predetermined threshold value, match the frequency peaks and frequency values in the frequency subset of the new inmate communication with a second subset of frequency data stored in a second detected record in the inmate communication database.

20. The system of claim 15, wherein the first and second voice biometric samples comprise audio voiceprints of the first and second inmates, and wherein the voice biometrics module stores a list of inmate identities corresponding to the audio voiceprints.