Selective security masking within recorded speech

• US 9,336,409 B2
• Filed: 08/07/2009
• Issued: 05/10/2016
• Est. Priority Date: 06/28/2006
• Status: Active Grant

• Current Assignee: Avaya LLC (Avaya Incorporated), Avaya Management L.P. (Avaya Incorporated)
• Original Assignee: Intellisist, Inc. dba Spoken Communications (Avaya Incorporated)
• Inventors: Doren, G. Kevin
• Primary Examiner(s): Coppola, Jacob C.
• Application Number:
  US12/462,750
• Publication Number:
  US 20090307779A1
• Time in Patent Office: 2,468 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 21/6263 during internet communication, e.g. revealing personal data from cookies
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656

Abstract

A marker is derived from an interaction between a person and an agent of a business and the agent'"'"'s user interface. A part of a speech signal that corresponds to a portion of the person'"'"'s special information is located with the marker. The speech signal results from the interaction between the person and the agent. The part of the speech signal that corresponds to the portion of the person'"'"'s special information is rendered unintelligible.

Claims

1. A method comprising:

monitoring a call comprising speech signals between a person and an agent of a business entity using a time base;

selecting an area of a user interface that accepts special information from the person during the call;

designating at least two offset distances around the area;

identifying an activity comprising at least one of a key stroke and mouse position in at least one of the offset distances and identifying a time associated with the activity using a further time base;

measuring a time offset between the time base and the further time base;

determining a correlation coefficient for a correlation between the activity and possible entry of the special information in the area based on a location of the activity in at least one of the offset distances in comparison with a location of the area;

applying a threshold to the correlation coefficient;

determining the entry of the special information and identifying a segment of the speech signals that corresponds to the special information within the call when the correlation coefficient satisfies the threshold using the time and the measured time offset; and

rendering at least a portion of the person'"'"'s special information unintelligible.

2. The method of claim 1, wherein the selection is based on a use case.

3. The method of claim 2, wherein the use-case is selected from the group consisting of sale of goods, and provider of services.

4. The method of claim 3, wherein a type of goods is selected from the group consisting of retail consumer goods, parts, wholesale goods user defined, and real estate.

5. The method of claim 3, wherein a type of services is selected from the group consisting of travel, medical, entertainment, after sales support, pre-sales support, user defined, and real estate.

6. The method of claim 1, further comprising:

monitoring the selected area of the user interface to obtain information on the activity.

7. The method of claim 6, wherein information on the activity in the area includes a time corresponding to the activity in the area.

8. The method of claim 1, wherein the special information is rendered unintelligible by encrypting the speech signal segment.

9. The method of claim 1, wherein the special information is rendered unintelligible by masking the speech signal segment.

10. The method of claim 1, wherein the special information is rendered unintelligible by deleting the speech signal segment.

11. The method of claim 1, wherein the speech signal is in a form of a voice recording.

12. The method of claim 11, further comprising:

analyzing the voice recording to provide feedback to the business based on the interaction.

13. The method of claim 12, wherein the feedback is in a form selected from the group consisting of print, audio, video, electronic, and multimedia.

14. The method of claim 1, wherein the speech signals are received as a data input apart from a voice recording.

15. The method of claim 14, wherein the data input is of a type selected from the group consisting of real-time, nearly real-time, and time delayed.