Speech Masking and Cancelling and Voice Obscuration

• US 20120053931A1
• Filed: 02/01/2011
• Published: 03/01/2012
• Est. Priority Date: 08/24/2010
• Status: Active Grant

• Current Assignee: Lawrence Livermore National Security LLC (Government of the United States of America)
• Original Assignee: Lawrence Livermore National Security LLC (Government of the United States of America)
• Inventors: Holzrichter, John F.
• Granted Patent: US 8,532,987 B2
• Time in Patent Office: Days
• US Class Current: 704/200.1
• CPC Class Codes: G10K 11/1754 Speech masking
• CPC Class Codes: G10L 15/28 Constructional details of speech recognition systems
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: H04K 2203/12 for acoustic communication
• CPC Class Codes: H04K 3/228 Elimination in the received signal of jamming or of data corrupted by jamming interference suppression in receivers H04B1/10
• CPC Class Codes: H04K 3/43 characterized by the control of the jamming power, signal-to-noise ratio or geographic coverage area
• CPC Class Codes: H04K 3/45 characterized by including monitoring of the target or target signal, e.g. in reactive jammers or follower jammers for example by means of an alternation of jamming phases and monitoring phases, called "look-through mode"
• CPC Class Codes: H04K 3/46 characterized in that the jamming signal is produced by retransmitting a received signal, after delay or processing
• CPC Class Codes: H04K 3/825 by jamming
• CPC Class Codes: H04R 3/002 Damping circuit arrangements for transducers, e.g. motional feedback circuits

Abstract

A non-acoustic sensor is used to measure a user'"'"'s speech and then broadcasts an obscuring acoustic signal diminishing the user'"'"'s vocal acoustic output intensity and/or distorting the voice sounds making them unintelligible to persons nearby.

Claims

1. An apparatus for measuring properties of a user'"'"'s speech production and user'"'"'s acoustic speech emissions, including a processor for said apparatus signals using an algorithm, and then broadcasting an obscuring acoustic signal diminishing the user'"'"'s vocal acoustic output intensity or distorting the voice sounds or both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons nearby, wherein the user has apparatus on or near the neck or head skin tissue, comprising:

one or more non-acoustic sensors proximate or contacting the user'"'"'s neck or head skin-tissue for sensing speech production and producing speech-signal timing information, speech type identification information, and apparatus control signals for purposes of producing an obscuring and or canceling acoustic signal, and

one or more acoustic sensors for measuring acoustic speech output of said user, for measuring the output of a loudspeaker together with user acoustic speech, and for measuring background noise, and

a processor to process signals from said non-acoustic and said acoustic sensors, which employs algorithms for purposes of real time generation and broadcasting of obscuring and/or canceling acoustic signals appropriate to the types of speech being spoken by user, and also to remove noise entering acoustic sensors, and also to store in a memory prior and instant segments of both non-acoustic and acoustic speech signals, as well as obscured speech signals, and background noise signals, and

a loudspeaker for generating acoustic signals for the purpose of diminishing the user'"'"'s vocal acoustic output intensity, or for distorting the voice sounds, or for both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons and instruments nearby, and

a second loudspeaker for purposes of transmitting the user'"'"'s acoustic speech, as detected by an acoustic sensor and after noise is removed, into the user'"'"'s ear or ears.

2. The apparatus of claim 1 wherein said non-acoustic sensor is a low-power electromagnetic radar.

3. The apparatus of claim 1 wherein said non-acoustic sensor is an electromagnetic interferometer with antenna.

4. The apparatus of claim 1 wherein said non-acoustic sensor is an accelerometer in contact with the user'"'"'s neck or head skin tissue.

5. The apparatus of claim 1 wherein said non-acoustic sensor is a laser interferometer vibration sensor.

6. The apparatus of claim 1 wherein said non-acoustic sensor is a partially coherent optical vibration sensor.

7. The apparatus of claim 1 wherein said non-acoustic sensor is a microphone sensor in contact with the user'"'"'s skin surface.

8. The apparatus of claim 1 wherein said non-acoustic sensor is a microphone proximate to user'"'"'s neck skin, measuring skin vibrations traversing a short air path to said sensor.

9. The apparatus of claim 1 wherein said non-acoustic sensor is an electric field sensor that detects muscle electrical signals associated with motions of speech production tissues.

10. The apparatus of claim 1 wherein said non-acoustic sensor is an electronic camera that detects lip, jaw, and tongue motions.

11. The apparatus of claim 1 wherein said apparatus includes an electronic signal processor that uses an output signal from non-acoustic sensor proximate to neck, a second output signal from acoustic sensor proximate to user'"'"'s mouth, and identifies a time period in advance of user'"'"'s speech leaving user'"'"'s mouth and nose, then during identified time period said processor generates and broadcasts a simultaneous obscuring and/or canceling acoustic signal to prevent user'"'"'s speech from being overheard by nearby listeners or instruments

12. An apparatus for measure a user'"'"'s speech and broadcasting an obscuring acoustic signal diminishing the user'"'"'s vocal acoustic output intensity or distorting the voice sounds or both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons nearby wherein the user has neck or head skin tissue, comprising:

a non-acoustic sensor proximate or contacting the user'"'"'s neck or head skin tissue for sensing speech production information,

a processor using said speech production information for producing an obscuring acoustic signal, and

a device for broadcasting said obscuring acoustic signal diminishing the user'"'"'s vocal acoustic output intensity or distorting the voice sounds or both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons nearby.

13. The apparatus of claim 12 wherein said non-acoustic sensor is a low-power electromagnetic radar.

14. The apparatus of claim 12 wherein said non-acoustic sensor is an electromagnetic interferometer with antenna.

15. The apparatus of claim 12 wherein said non-acoustic sensor is an accelerometer in contact with the user'"'"'s neck or head skin tissue.

16. The apparatus of claim 12 wherein said non-acoustic sensor is a laser interferometer vibration sensor.

17. The apparatus of claim 12 wherein said non-acoustic sensor is a partially coherent optical vibration sensor.

18. The apparatus of claim 12 wherein said non-acoustic sensor is a microphone in contact with the user'"'"'s skin surface.

19. The apparatus of claim 12 wherein said non-acoustic sensor is an electric field sensor that detects muscle electrical signals associated with motions.

20. A method of measuring both a user'"'"'s speech and also measuring a user'"'"'s speech production for voiced speech time periods, and then broadcasting an obscuring acoustic signal diminishing the user'"'"'s vocal acoustic output intensity or distorting the voice sounds or both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons or instruments nearby, comprising the steps of:

using a non-acoustic sensor proximate to or contacting the user'"'"'s neck or head for sensing speech production, using an acoustic sensor to measure and record user'"'"'s emitted speech signals, then using said non-acoustic sensor signals and said recordings of user'"'"'s prior acoustic speech signals, calculating an obscuring and/or canceling signal, then broadcasting said calculated signal as an acoustic signal diminishing the user'"'"'s vocal acoustic output intensity or distorting the voice sounds or both diminishing the user'"'"'s vocal acoustic output intensity and distorting the voice sounds making them unintelligible to persons or sound sensing instruments nearby.

21. The method of claim 20 for obtaining the moment in future time when obscuring and canceling acoustic emissions are broadcast, whereas said future time is estimated by adding a prior determined delay-time-value to the time in the present voiced-speech period when a non-acoustic sensor measures the fastest negative-going signal of vocal fold closure, and whereas said prior determined delay-time-value was obtained from data recorded during a prior speech time period, during which the start time in said prior time period was defined as the time of the negative vocal fold closure signal and the end time in said prior time period was defined as the corresponding time of the negative acoustic signal, the difference between the prior start time and the prior end time gives said prior determined delay-time-value.

22. The method of claim 20 for identifying and then canceling and obscuring an unvoiced speech segment by using non-acoustic and acoustic sensors to identify the onset of said unvoiced speech period in a time period short compared to its duration, then using a processor to select the appropriate signal to transmit to a loudspeaker, including the choice of transmitting a pink noise, selected noise, or determinate noise signal, and then broadcasting said selected noise-like signal for the rest of the unvoiced speech time period and into the following time period of voiced or silent speech as needed for best obscuration.

23. The method of claim 20 for identifying and then canceling and obscuring a silent speech segment by using non-acoustic and acoustic sensors to identify the onset of said silent speech period rapidly compared to its duration, then using a processor to select the appropriate signal to transmit to a loudspeaker, including the choice of transmitting a zero amplitude signal or a prior recorded noise background signal, then broadcasting a selected noise-like or silent signal for the rest of the silent speech time period.

24. The method of claim 20 and claim 21 including an algorithm that modifies the non-acoustic sensor signal by altering its amplitude, its phase, its amplitude vs frequency, its phase versus frequency, its delay, its polarity, and its format for broadcast at a determined time in the future by a loud speaker which causes a reduction in intensity and intelligibility as detected by a listener.

25. The method of claim 20 and claim 21 including an algorithm that modifies an acoustic sensor signal by altering its amplitude, its phase, its amplitude vs frequency, its phase versus frequency, its delay, its polarity, and its format for broadcast at a determined time in the future by a loud speaker which causes a reduced intensity and intelligibility, as detected by a listener.

26. The method of claim 22 and claim 23 such that the resulting signal as broadcast in the future by a loudspeaker both reduces the intensity and reduces the intelligibility for at least 3 kinds of human speech so that the speaker'"'"'s voice is reduced in intensity and obscured as detected by a listener or sound detecting instrument.

27. The method of claims 21, 22, 23, 24, and 25 in which a non-acoustic sensor and algorithms, in conjunction with prior acoustic signals from an acoustic speech sensor, are used to generate an obscuring signal, while an all-acoustic noise-canceling microphone and prior art noise reducing algorithms enable the user to detect his/her own undistorted or undiminished speech signal for subsequent use.

28. The method of claim 27 in which the output signal from a noise-canceling microphone and from noise reducing algorithms provide a signal sufficiently free of background noise, obscuring noise, canceling noise, and other system noise that said signal when conveyed to the user'"'"'s own ear or ears and to other instruments, has a low noise level and is highly intelligible as defined by the user.