Natural-language surveillance of packet-based communications

• US 8,059,790 B1
• Filed: 11/01/2006
• Issued: 11/15/2011
• Est. Priority Date: 06/27/2006
• Status: Active Grant

• Current Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Original Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Inventors: Acosta, Irwin Joseph Butac, Kunkel, Philip M., Paterik, Thomas L., Weaver, Farni B.
• Primary Examiner(s): PATEL, HEMANT SHANTILAL
• Application Number:
  US11/555,584
• Time in Patent Office: 1,840 Days
• Field of Search: 340 51- 525, 340 58- 585, 340/3.1, 340 33- 332, 340 35- 39, 370351-356, 370259-271, 379 671- 8828, 379/201.01, 379/207.13, 379/235, 379/265.07, 379/7, 379/38, 379/48, 379 9301- 9304, 379 9326- 9337, 379/101.01, 37910601-10608, 379/201.11, 379/211.01, 379/338, 379/339, 455410-417, 704246-257, 704270-278, 709201-203, 709/206, 709/207, 709217-219, 709224-229, 709/231, 709/232, 709/239, 709/240, 709/243, 709/244, 348 1401- 1416
• US Class Current: 379/88.02
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: H04L 63/1408 by monitoring network traffic monitoring network traffic per se H04L43/00
• CPC Class Codes: H04L 63/30 for supporting lawful interception, monitoring or retaining of communications or communication related information circuit switched telephony call monitoring H04M3/2281
• CPC Class Codes: H04L 65/764 at the destination reformatting of additional data in video clients H04N21/4355
• CPC Class Codes: H04L 67/30 Profiles
• CPC Class Codes: H04M 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 3/2281 Call monitoring, e.g. for law enforcement purposes; Call tracing; Detection or prevention of malicious calls

Abstract

Methods and systems are provided for natural-language surveillance of packet-based communications. Speech data is received, representing at least part of a speech session. The speech data is compared with at least one speech-characteristics profile in an automated manner. The comparing step comprises using natural-language-processing (NLP) to compare the speech data with the profiles on at least one of a lexical level, a syntactic level, a semantic level, a discourse level, and a pragmatic level. It is determined in an automated manner whether the speech data matches any of the profiles. If the speech data matches at least one profile, an indication that the speech data matches the at least one matching profile is responsively stored.

Claims

1. A method comprising:

a speech-analysis platform receiving speech data representing at least part of a speech session;

the speech-analysis platform comparing the speech data with each of at least one stored speech-characteristics profile in an automated manner, wherein each such profile is associated with a person that is a target of surveillance, wherein the comparing step comprises using natural-language-processing (NLP) to compare the speech data with the at least one stored speech-characteristics profile on each of a lexical level, a syntactic level, a semantic level, and a discourse level, wherein said using of NLP does not comprise conducting voiceprint analysis;

the speech-analysis platform determining in an automated manner from said using of NLP whether the speech data matches any of the profiles; and

if the speech-analysis platform determines from said using of NLP that the speech data matches at least one of the profiles on each of the lexical level, the syntactic level, the semantic level, and the discourse level, the speech-analysis platform responsively storing an indication that the speech data matches at least one of the profiles.

2. The method of claim 1, wherein the speech data was produced at least in part by at least one of a computer, a text-to-speech (TTS) technology, a speech-to-text (STT) technology, and a voice-alteration technology.

3. The method of claim 1, wherein the speech session comprises at least one of a telephone call, a voicemail message, an answering-machine message, a voice over Internet Protocol (VoIP) session, an instant-messaging (IM) session, a chat-room session, a message-board posting, and an e-mail.

4. The method of claim 1, wherein the profiles comprise respective word-choice data, and wherein using NLP to compare the speech data with the profiles on the lexical level comprises comparing at least one word represented by the speech data with the respective word-choice data of at least one profile.

5. The method of claim 1, wherein the profiles comprise respective syntax-choice data, and wherein using NLP to compare the speech data with the profiles on the syntactic level comprises comparing at least one instance of syntax usage represented by the speech data with the respective syntax-choice data of at least one profile.

6. The method of claim 1, wherein the profiles comprise respective semantic-usage data, and wherein using NLP to compare the speech data with the profiles on the semantic level comprises comparing at least one instance of semantic usage represented by the speech data with the respective semantic-choice data of at least one profile.

7. The method of claim 1, wherein the profiles comprise respective discourse-level data, wherein using NLP to compare the speech data with the profiles on the discourse level comprises comparing the speech data with the respective discourse-level data of at least one profile.

8. The method of claim 1, wherein the profiles comprise respective pragmatic-level data, the method further comprising using the respective pragmatic-level data of at least one profile to compare the speech data with the profiles on at least one of the lexical level, the syntactic level, the semantic level, and the discourse level.

9. The method of claim 1, wherein determining whether the speech data matches any of the profiles comprises determining whether the speech data matches any of the profiles to a predetermined degree of statistical significance.

10. The method of claim 1, further comprising conducting a voiceprint analysis of the speech data.

11. The method of claim 10, wherein the steps of claim 1 are carried out in response to a determination that the voiceprint analysis of the speech data has failed.

12. The method of claim 1, further comprising:

if the speech-analysis platform determines that the speech data matches at least one of the profiles on each of the lexical level, the syntactic level, the semantic level, and the discourse level, the speech-analysis platform responsively further carrying out at least one of recording the speech session, alerting a monitoring entity of the match between the speech data and the at least one matching profile, and forwarding a copy of the speech session to a monitoring entity.