Filtering data in an audio stream

• US 10,453,447 B2
• Filed: 11/28/2017
• Issued: 10/22/2019
• Est. Priority Date: 11/28/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Schmidt, Jeffrey A., Braham, Adam D., Nikolai, Jason A., Santosuosso, John M.
• Primary Examiner(s): Abebe, Daniel
• Application Number:
  US15/824,051
• Publication Number:
  US 20190164539A1
• Time in Patent Office: 693 Days
• Field of Search: 704250
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2203/6009 Personal information, e.g. profiles or personal directories being only provided to authorised persons
• CPC Class Codes: H04M 3/436 Arrangements for screening incoming calls , i.e. evaluating the characteristics of a call before deciding whether to answer it based on the calling party profile H04M3/42059; based on location H04M3/42348; based on presence H04M3/42365; diversion H04M3/54
• CPC Class Codes: H04M 3/51 Centralised call answering arrangements requiring operator intervention , e.g. call or contact centers for telemarketing
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends

Abstract

Disclosed herein is a system, method, and computer program product for filtering data in an audio stream. Incoming audio is converted into text and analyzed for information of a first type. Based on the detection of information of the first type from a source, a recipient is blocked from receiving the audible information.

Claims

1. A system comprising:

at least one processor;

at least one memory component;

a live audio stream having one or more spoken words,

wherein the live audio stream is spoken language during a telephone call;

a sensitive word list,

wherein the sensitive word list is based on credentials of a listener;

at least one speech-to-text engine configured to decode the live audio stream into text;

a sensitive word detecting engine configured to analyze the text for text strings indicating sensitive information upon matching words in the text strings to the sensitive word list; and

a blocking engine configured to block the live audio stream when words matching the sensitive word list are detected from the live audio stream.

2. The system of claim 1,

wherein the sensitive information is selected from a group consisting of financial information, health information, credit information, passcode information, identity information, and location information;

and

wherein the sensitive information is included in the sensitive word list; and

wherein the detecting is based upon an incomplete portion of the sensitive information that has been spoken before the detecting is complete.

3. The system of claim 1, wherein the analyzing comprises comparing the text with a known list of sensitive trigger words.

4. The system of claim 1, wherein the text strings comprise one or more words that commonly come before sensitive information.

5. The system of claim 1, wherein the text strings comprise sensitive information.

6. The system of claim 1 further comprising selecting a speech-to-text engine that meets or exceeds an accuracy threshold.

7. A method for blocking sensitive information in an live audio stream comprising:

receiving the live audio stream, the live audio stream having spoken words, from a live audio feed;

converting the live audio stream into text with at least one speech-to-text engine;

analyzing, by a sensitive word detection engine, the text for text strings indicating sensitive information upon matching words in the text strings to a sensitive word list,

wherein the sensitive word list is based on credentials of a listener; and

blocking, by a blocking engine, at least a portion of the live audio stream wherein the portion of the live audio stream that is blocked is at least a portion of the detected sensitive words.

8. The method of claim 7, wherein the live audio stream is a telephone call.

9. The method of claim 7, wherein the sensitive information is selected from a group consisting of financial information, health information, credit information, passcode information, identity information, and location information.

10. The method of claim 7, wherein the analyzing comprises comparing the text with a known list of sensitive trigger words.

11. The method of claim 7, wherein the text strings comprise one or more words that commonly come before sensitive information.

12. The method of claim 7, wherein the text strings comprise sensitive information.

13. The method of claim 7 further comprising selecting a speech-to-text engine that meets or exceeds an accuracy threshold.

14. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:

receiving the live audio stream, the live audio stream having spoken words, from a live audio feed;

converting the live audio stream into text with at least one speech-to-text engine;

analyzing, by a sensitive word detection engine, the text for text strings indicating sensitive information upon matching words in the text strings to a sensitive word list,

wherein the sensitive word list is based on credentials of a listener; and

blocking, by a blocking engine, at least a portion of the live audio stream wherein the portion of the live audio stream that is blocked is at least a portion of the detected sensitive words.

15. The computer program product of claim 14, wherein the live audio stream is a telephone call.

16. The computer program product of claim 14, wherein the analyzing comprises comparing the text with a known list of sensitive trigger words.

17. The computer program product of claim 14, wherein the text strings comprise one or more words that commonly come before sensitive information.

18. The computer program product of claim 14, wherein the text strings comprise sensitive information.