Determining characteristics of received voice data packets to assist prosody analysis

• US 20040059935A1
• Filed: 08/29/2003
• Published: 03/25/2004
• Est. Priority Date: 10/19/2001
• Status: Active Grant

• Current Assignee: Raytheon BBN Technlogies Corp. (Rtx Corporation)
• Original Assignee: BBN Technologies (Rtx Corporation)
• Inventors: Cousins, David Bruce
• Granted Patent: US 7,263,479 B2
• Time in Patent Office: Days
• US Class Current: 713/200
• CPC Class Codes: H04L 63/045 wherein the sending and receiving network entities apply hybrid encryption, i.e. combination of symmetric and asymmetric encryption cryptographic mechanisms or cryptographic arrangements using a plurality of keys or algorithms H04L9/14
• CPC Class Codes: H04L 63/1408 by monitoring network traffic monitoring network traffic per se H04L43/00

Abstract

A method and system are provided for acquiring information about communication among nodes [110, 210] in a network [100, 200] by intercepting chunks of data in the network by a tap [120, 220] located among the nodes [110, 210]. A file [740] of data, including characteristics [400] of the intercepted chunks may be produced. The data may be converted into at least one time series and processed to produce prosody information. The prosody information may be used by prosody analysis.

Claims

1. A method of acquiring information about communication among nodes in a network and producing data to be used to derive prosody information, the method comprising:

intercepting chunks of data in the network by a tap located among the nodes in the network;

generating a file of data from the intercepted ones of the chunks of data, the data including characteristics of the intercepted chunks of data;

converting the data into at least one time series; and

processing the at least one time series to produce prosody information.

2. The method of claim 1, further comprising eliminating any of the data corresponding to undesired ones of the intercepted chunks of data.

3. The method of claim 2, wherein the generated file of data includes at least one of a source, a destination, a protocol identifier, a time of arrival and a size of each of the intercepted chunks of data.

4. The method of claim 3, wherein the generated file serves as input for eliminating any of the data corresponding to undesired ones of the intercepted chunks of data.

5. The method of claim 1, wherein the chunks of data include voice over IP data.

6. The method of claim 1, wherein the chunks of data include encrypted speech data.

7. The method of claim 1, wherein the processing comprises producing an altered time series as the prosody information, the altered time series showing a duration of at least one utterance and a duration of at least one period of silence.

8. The method of claim 1, wherein the intercepting includes executing a TCP dump to intercept the chunks of data detected by the tap.

9. The method of claim 1, wherein the processing includes processing each of the at least one time series separately and computing a mean for each of the at least one time series.

10. The method of claim 9, wherein each of the at least one time series corresponds to communications from a respective first node to a respective second node.

11. The method of claim 9, wherein the mean for each of the at least one time series is calculated every specific time period using data from a most current of the specific time periods of data of a respective one of the at least one time series.

12. The method of claim 1, wherein the network is a wireless network and the tap is configured to intercept wireless communications.

13. The method of claim 1, wherein the network is a wired network and the tap is configured to intercept communications on the wired network.

14. An apparatus in a network, comprising:

means for monitoring chunks of data in the network;

means for generating data from the monitored chunks of data, the data including characteristics of the monitored chunks of data;

means for converting the generated data into at least one time series; and

means for processing the at least one time series to produce prosody information for prosody analysis.

15. The apparatus of claim 14, further comprising means for eliminating at least a portion of the data corresponding to undesired ones of the chunks of data.

16. The apparatus of claim 14, wherein the means for processing comprises means for producing an altered time series as the prosody information, the altered time series showing a duration of at least one utterance and a duration of at least one period of silence.

17. A computer-readable medium that stores instructions executable by one or more processors to perform a method for acquiring information about communication among nodes in a network and for producing prosody information, comprising:

instructions for receiving chunks of data in the network;

instructions for generating characteristic data from the received chunks of data;

instructions for producing at least one time series from the characteristic data;

instructions for processing the at least one time series to produce prosody information; and

instructions for outputting the prosody information.

18. The computer-readable medium of claim 17, further comprising instructions for eliminating characteristic data corresponding to undesired ones of the received chunks of data.

19. The computer-readable medium of claim 17, wherein the instructions for processing comprises instructions for producing an altered time series as the prosody information, the altered time series showing a duration of at least one utterance and a duration of at least one period of silence.

20. The computer-readable medium of claim 17, wherein the characteristic data includes at least one of a source, a destination, a protocol, a time of arrival and a size of each of the chunks of data.

21. The computer-readable medium of claim 17, wherein the instructions for processing the at least one time series further comprise instructions for processing each of the at least one time series separately and for computing a mean for each of the at least one time series.

22. The computer-readable medium of claim 21, wherein the instructions for computing a mean further include instructions for calculating the mean of one of the at least one time series using data from a most recent specific time period of data from the respective one of the at least one time series.

23. A system for acquiring information about communication among nodes in a network and producing data to be used to derive prosody information, comprising:

a monitoring component to monitor data communications in the network;

a data generating component to generate data, including characteristics of the monitored data communications;

a time series producing component to convert the generated data into at least one time series; and

a processing component to process the at least one time series to produce prosody information for prosody analysis.