Method and apparatus for speech privacy

• US 20060247919A1
• Filed: 01/04/2006
• Published: 11/02/2006
• Est. Priority Date: 01/10/2005
• Status: Active Grant

• Current Assignee: Herman Miller Incorporated (Millerknoll Incorporated)
• Original Assignee: Herman Miller Incorporated (Millerknoll Incorporated)
• Inventors: Mapes-Riordan, Daniel, Specht, Jeffrey, DeKruif, William
• Granted Patent: US 7,376,557 B2
• Time in Patent Office: Days
• US Class Current: 704/201
• CPC Class Codes: G10L 21/06 Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids G10L15/26 takes precedence
• CPC Class Codes: H04K 1/02 by adding a second signal to make the desired signal unintelligible
• CPC Class Codes: H04K 1/10 by using two signals transmitted simultaneously or successively
• CPC Class Codes: H04K 2203/12 for acoustic communication
• 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 3/825 by jamming

Abstract

A privacy apparatus adds a privacy sound based on a speaker'"'"'s own voice into the environment, thereby confusing listeners as to which of the sounds is the real source. This permits disruption of the ability to understand the source speech of the user by eliminating segregation cues that the auditory system uses to interpret speech. The privacy apparatus minimizes segregation cues. The privacy apparatus is relatively quiet and thus easily acceptable in a typical open floor design office space. The privacy apparatus contains an A/D converter that converts the speech into a digital signal, a DSP that converts the digital signal into a privacy signal with pre-recorded speech fragments of the person speaking, a D/A converter that converts the privacy signal into an output signal and one or more loudspeakers from which the output signal is emitted.

Claims

1. A method of disrupting speech of at least one talker emanating from a space, the method comprising:

accessing a plurality of speech signals in a memory;

generating at least one privacy output signal comprised of the speech signals being summed with one another so that the speech signals at least partly overlap one another; and

outputting the at least one privacy output signal in order to disrupt the speech emanating from the space.

2. The method of claim 1, where generating at least one privacy output signal comprises:

generating plurality of voice streams, each voice stream being generated by selecting at least some of the speech signals in the memory and assembling the selected speech signals into the voice stream

3. The method of claim 2, where the speech signals are based on speech from the talker.

4. The method of claim 3, where the speech signals comprise phonemes.

5. The method of claim 3, further comprising inputting speech from the talker during a training mode;

and

selecting speech fragments from the input speech; and

storing the speech fragments in the memory.

6. The method of claim 5, where selecting speech fragments from the input speech comprises:

determining an increase in energy level of the input speech as a beginning of the speech fragment; and

determining a decrease in the energy level of the input speech as an end of the speech fragment.

7. The method of claim 5, further comprising:

selecting a first set of speech fragments from the input speech based on first predetermined criteria;

selecting a second set of speech fragments from the input speech based on second predetermined criteria;

wherein generating a plurality of voice streams comprises generating at least one voice stream from the first set of speech fragments and generating at least one voice streams from the second set of speech fragments.

8. The method of claim 1, where the speech signals are based on speech from someone other than the talker.

9. The method of claim 2, where the plurality of voice streams are uncorrelated with one another.

10. The method of claim 9, where the speech signals comprises speech fragments of the talker;

and

where generating a plurality of voice streams comprises, for each voice stream, randomly selecting speech fragments from the memory.

11. The method of claim 2, where the plurality of voice streams are generated by assembling the selected speech signals along with gaps into the voice stream, the gaps being sections that comprise no speech signals.

12. The method of claim 11, where time lengths of the gaps are selected within a predefined range.

13. The method of claim 11, where the time lengths of the gaps are randomly selected within the predefined range.

14. The method of claim 1, where the speech of the at least one talker comprises speech into a telephone handset;

and

further comprising;

sensing loudness of the speech of the at least one talker into the telephone handset; and

determining loudness of the at least one privacy output signal based on the loudness of the speech of the at least one talker into the telephone handset.

15. The method of claim 2, where speech from a plurality of talkers emanates from the space;

where the memory comprises speech fragments from each of the plurality of talkers; and

where generating a plurality of voice streams comprises generating multiple voice streams for each the plurality of talkers.

16. The method of claim 15, further comprising identifying at least one of the plurality of talkers;

and

where accessing a plurality of speech signals in a memory comprises selecting a set of speech fragments based on identifying of at least one of the talkers.

17. An apparatus for disrupting speech of at least one talker emanating from a space, the apparatus comprising:

a microphone that receives a voice of a person speaking;

a processor that generates a privacy signal, the privacy signal comprised of the fragments of the voice received by the microphone being summed with one another so that the fragments at least partly overlap one another, and

at least one loudspeaker for emitting the privacy signal to disrupt the speech of the talker emanating from the space.

18. The apparatus of claim 17, where the privacy signal is comprised of a plurality of voice streams based on the voice received by the microphone

19. The apparatus of claim 18, wherein the plurality of voice streams comprise speech fragments selected from the voice received by the microphone.

20. The apparatus of claim 19, where the processor generates the speech fragments;

and

further comprising a memory for storing the speech fragments.

21. The apparatus of claim 20, where the processor randomly selects the speech fragments to create the plurality of voice streams from the speech fragments stored in the memory.

22. The apparatus of claim 21, where the speech fragments comprise fragments that exhibit characteristics of phonemes.

23. The apparatus of claim 18, where the loudspeaker includes input from a first channel and a second channel;

where the processor sums a plurality of voice streams for input on the first channel of the loudspeaker and sums a different plurality of voice streams for input on the second channel of the loudspeaker.

24. The apparatus of claim 23, where the plurality of voice streams are uncorrelated with one another.

25. The apparatus of claim 17, where the processor receives input from a telephone regarding a level of loudness of speech input to the telephone;

and

where the processor determines a level of output for the privacy signal based on the level of loudness.

26. The apparatus of claim 25, where the telephone comprises a mouthpiece that covers, at least partly, a mouth region of the talker.

27. The apparatus of claim 18, where the privacy signal comprises a plurality of voice streams composed of speech fragments based on the voice received by the microphone and gaps comprising no speech signals.

28. The apparatus of claim 18, where speech from a plurality of talkers emanates from the space;

further comprising a memory comprising speech fragments from each of the plurality of talkers; and

where generating a plurality of voice streams comprises generating multiple voice streams for each the plurality of talkers.

29. The apparatus of claim 28, where the processor identifies at least one of the plurality of talkers;

and

where the processor selects a set of speech fragments based on identifying of at least one of the talkers.