Trusted speech transcription

• US 10,621,977 B2
• Filed: 10/30/2015
• Issued: 04/14/2020
• Est. Priority Date: 10/30/2015
• Status: Active Grant

• Current Assignee: McAfee, LLC
• Original Assignee: McAfee, LLC
• Inventors: Pogorelik, Oleg, Nayshtut, Alex, Muttik, Igor, Shaliv, Adi
• Primary Examiner(s): Jackson, Jakieda R
• Application Number:
  US14/928,749
• Publication Number:
  US 20170125014A1
• Time in Patent Office: 1,628 Days
• Field of Search: 704235
• CPC Class Codes: G06F 16/61 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/685 using automatically derived transcript of audio data, e.g. lyrics speech recognition G10L15/00
• CPC Class Codes: G06F 21/101 by binding digital rights to specific entities
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G06F 2221/2107 File encryption
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase
• CPC Class Codes: G10L 25/51 for comparison or discrimination

Abstract

Embodiments include a system, method, and apparatus for creating a trusted speech transcription. Transcription logic can receive a signal of audible speech from an audio source and convert the audible speech signal into text. Sampling logic can receive the signal of the audible speech and record a portion of the of the audible speech as a digital audio recording of the audible speech. Packaging logic can create a trusted speech transcription data record that includes the text and the digital audio recording and to digitally sign the transcription file to securely associate the text with the digital audio recording.

Claims

1. A trusted speech transcription apparatus for creating a trusted speech transcription data record, the apparatus comprising:

transcription logic implemented at least partially in hardware to receive a signal of audible speech from an audio source and to convert the signal of audible speech into text;

sampling logic implemented at least partially in hardware to receive the signal of the audible speech and to record the signal of the audible speech as a digital audio recording of the audible speech;

packaging logic implemented at least partially in hardware in a trusted execution environment to identify a speaker of the audible speech and to create the trusted speech transcription data record, wherein the trusted speech transcription data record is a single file that includes a portion of the text, a corresponding portion of the digital audio recording, the identity of the speaker, and a digital signature that uses a digital encryption key; and

activation logic to receive the signal of audible speech and to activate the transcription logic and the sampling logic upon detecting one or more of;

a predetermined word or phrase within the signal of audible speech;

a signal quality of the signal of audible speech satisfying a clarity parameter;

an identification of a specific speaker of the audible speech, from the signal of audible speech;

or

a duration of the audible speech exceeds a threshold duration.

2. The apparatus of claim 1, wherein the packaging logic further tags the trusted speech transcription data record with metadata associated with the audible speech.

3. The apparatus of claim 2, wherein the metadata comprises one or more of a time stamp, an internet protocol (IP) address of the audio source, an identification of a speaker of the audible speech, or an identifier of the source of the audible speech.

4. The apparatus of claim 1, wherein the transcription logic converts the signal of the audible speech and the sampling logic records the signal of audible speech in parallel.

5. The apparatus of claim 1, wherein the sampling logic records a fragment of the signal of audible speech and the digital audio recording is a recording of the fragment of the audible speech, the fragment of the audible speech representative of a source of the audible speech, and wherein the trusted speech transcription data record further includes the fragment of the audible speech.

6. The apparatus of claim 5, wherein the fragment of the audible speech comprises quality characteristics to identify a speaker of the audible speech.

7. The apparatus of claim 5, wherein the fragment of the audible speech is stored with a corresponding transcription of the audible speech.

8. A computer program product tangibly embodied on a non-transitory computer readable medium, the computer program product including instructions that, when executed by a processing element causes the processing element to:

receive an electronic signal of audible speech from an audio source;

upon detecting one or more of;

a predetermined word or phrase within the electronic signal of audible speech;

a signal quality of the electronic signal of audible speech satisfying a clarity parameter;

an identification of a specific speaker of the audible speech, from the electronic signal of audible speech;

or

a duration of the audible speech exceeds a threshold duration;

convert the electronic signal of audible speech into digital text; and

record the electronic signal of audible speech as a digital audio recording of the audible speech;

identify a speaker of the audible speech; and

create, in a trusted execution environment, a trusted speech transcription data record, wherein the trusted speech transcription data record is a single file that includes the digital text, the digital audio recording, the identity of the speaker, and a digital signature.

9. The computer program product of claim 8, wherein instructions are further operable to tag the trusted speech transcription data record with metadata associated with the audible speech.

10. The computer program product of claim 9, wherein the metadata comprises one or more of a time stamp, an internet protocol (IP) address of the audio source, an identification of a speaker of the audible speech, or an identifier of the source of the audible speech.

11. The computer program product of claim 8, wherein the instructions are further operable to convert the electronic signal of the audible speech and to record the electronic signal of audible speech in parallel.

12. The computer program product of claim 8, wherein instructions are further operable to record a fragment of the electronic signal of audible speech and the digital audio recording is a recording of the fragment of the audible speech, the fragment of the audible speech representative of a source of the audible speech, and wherein the trusted speech transcription data record further includes the fragment of the audible speech.

13. The computer program product of claim 12, wherein the fragment of the audible speech comprises quality characteristics to identify a speaker of the audible speech.

14. The computer program product of claim 12, wherein the fragment of the audible speech is stored with a corresponding transcription of the audible speech.

15. A method comprising:

receiving an electronic signal of audible speech from an audio source;

identifying at least a portion of the audible speech that includes identification cues of the identity of a speaker of the audible speech;

upon detecting one or more of;

a predetermined word or phrase within the electronic signal of audible speech;

a signal quality of the electronic signal of audible speech satisfying a clarity parameter;

an identification of a specific speaker of the audible speech, from the electronic signal of audible speech;

or

a duration of the audible speech exceeds a threshold duration;

converting the electronic signal into digital text; and

recording the electronic signal of audible speech as a digital audio recording of the audible speech; and

creating, in a trusted execution environment, a trusted speech transcription data record, wherein the trusted speech transcription data record is a single file that includes the digital text, the digital audio recording, the identity of the speaker, and a digital signature.

16. The method of claim 15, wherein recording the electronic signal of audible speech comprises recording a fragment of the audible speech, wherein the digital audio recording is a recording of the fragment of the audible speech, wherein the fragment of the audible speech is representative of a source of the audible speech and corresponds to the transcribed speech, and wherein the trusted speech transcription data record further includes the fragment of the audible speech.