Methods, apparatus, and articles of manufacture to generate voices for artificial speech based on an identifier represented by frequency dependent bits

• US 10,468,013 B2
• Filed: 03/31/2017
• Issued: 11/05/2019
• Est. Priority Date: 03/31/2017
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Pogorelik, Oleg
• Primary Examiner(s): Kazeminezhad, Farzad
• Application Number:
  US15/476,421
• Publication Number:
  US 20180286384A1
• Time in Patent Office: 949 Days
• Field of Search: 700245, 700246, 726 26
• CPC Class Codes: A47L 5/30 with driven dust-loosening tools, e.g. rotating brushes
• CPC Class Codes: G06F 21/10 Protecting distributed programs or content, e.g. vending or licensing of copyrighted material protection in video systems or pay television H04N7/16 ; Digital rights management [DRM]
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser

Abstract

Methods, apparatus, and articles of manufacture to generate voices for artificial speech are disclosed. An example apparatus includes a component storing an identifier, the identifier uniquely identifying the apparatus from a plurality of apparatus, an artificial speech generator to generate a first artificial speech signal representing text, the first artificial speech signal generated based on the identifier which is represented by frequency dependent bits by assigning specific bits to specific frequency bands, the first artificial speech signal audibly different from artificial speech signals generated by respective ones of the plurality of apparatus for the text, and an output device to output an audible signal representing the first artificial speech signal.

Claims

1. An apparatus comprising:

an artificial speech generator to;

mix a first artificial speech signal with a voice personalization signal to generate a modified artificial speech signal, the first artificial speech signal representing text, the voice personalization signal corresponding to a subset of sounds associated with the first artificial speech signal, the subset of sounds corresponding to a subset of frequency bands in a frequency-domain representation of the first artificial speech signal;

identify the subset of sounds based on an identifier, the identifier to uniquely identify the apparatus from a plurality of other apparatus, the identifier defined by a string of bits; and

assign individual bits of the string of bits to different ones of the frequency bands, the subset of the frequency bands corresponding to the subset of sounds identified based on values of the individual bits; and

an output device to output an audible signal based on the modified artificial speech signal, the audible signal to be audibly different to a human listener from other audible signals generated by respective ones of the plurality of other apparatus for the same text.

2. The apparatus as defined in claim 1, wherein the artificial speech generator includes:

a text-to-speech converter to convert the text to generate the first artificial speech signal;

a voice personalizer to modify the first artificial speech signal using the identifier to form the voice personalization signal; and

a modifier to mix the first artificial speech signal with the voice personalization signal to generate the modified artificial speech signal.

3. The apparatus as defined in claim 2, wherein the modifier includes an audio mixer to combine the first artificial speech signal and the voice personalization signal to generate the modified artificial speech signal.

4. The apparatus as defined in claim 3, wherein the audio mixer is at least one of a digital mixer, a digital audio output signal or an analog mixer.

5. The apparatus as defined in claim 1, wherein the artificial speech generator includes:

a text-to-speech converter to convert the text to generate the first artificial speech signal; and

a modifier to mix the first artificial speech signal with the voice personalization signal to generate the modified artificial speech signal.

6. The apparatus as defined in claim 1, wherein the subset of sounds includes at least one of a harmonic, or a disharmonious overtone.

7. The apparatus as defined in claim 1, further including;

a processor to execute an operating system; and

an application programming interface, the artificial speech generator to obtain the identifier using the application programming interface.

8. The apparatus as defined in claim 2, wherein the voice personalizer includes:

a frequency-domain representation generator to generate the frequency-domain representation of the first artificial speech signal;

a frequency-domain identifier generator to generate a frequency-domain identifier from the identifier;

a frequency-domain modifier to modify the frequency-domain representation with the frequency-domain identifier to form a frequency-domain voice personalization signal; and

a frequency-to-time domain converter to form the voice personalization signal from the frequency-domain voice personalization signal.

9. The apparatus as defined in claim 1, wherein the output device is at least one of a speaker, an audio jack, or a digital-to-analog converter.

10. The apparatus as defined in claim 1, further including:

a non-transitory computer-readable storage medium including instructions; and

processor circuitry to execute the instructions to generate the modified artificial speech signal.

11. The apparatus as defined in claim 1, wherein the artificial speech generator includes:

a frequency-domain representation generator to generate the frequency-domain representation of the first artificial speech signal, the frequency-domain representation to include different frequencies; and

a frequency-domain modifier to modify individual ones of the different frequencies based on values of corresponding individual bits of the plurality of bits defining the identifier.

12. The apparatus as defined in claim 1, wherein the identifier includes a serial number.

13. The apparatus as defined in claim 1, wherein the identifier includes an international mobile equipment identity (IMEI) number.

14. The apparatus as defined in claim 1, wherein the identifier includes a media access control (MAC) address.

15. The apparatus as defined in claim 1, wherein the identifier includes a device unique identifier (DUID).

16. The apparatus as defined in claim 1, wherein the identifier includes a processor serial number.

17. The apparatus as defined in claim 1, wherein the identifier includes a central processing unit (CPU) serial number.

18. The apparatus as defined in claim 1, wherein the identifier includes a management information base (MIB) identifier.

19. The apparatus as defined in claim 1, wherein the identifier includes a Bluetooth®

address.

20. The apparatus as defined in claim 1, wherein the subset of the frequency bands includes a first set of the frequency bands and excludes a second set of the frequency bands, the individual bits having a first value assigned to the first set of the frequency bands, the individual bits having a second value assigned to the second set of frequency bands.

21. The apparatus as defined in claim 1, wherein the frequency bands extend over a range of frequencies associated with audible speech.

22. A method comprising:

converting text to an initial artificial speech signal;

generating a voice personalization signal corresponding to a subset of sounds associated with the initial artificial speech signal, the subset of sounds corresponding to a subset of frequency bands in a frequency-domain representation of the initial artificial speech signal, the subset of the frequency bands selected based on an identifier, the identifier defined by a string of bits, the subset of the frequency bands corresponding to the subset of sounds identified based on values of individual bits of the string of bits assigned to different ones of the frequency bands;

combining the initial artificial speech signal with the voice personalization signal to generate a modified artificial speech signal, the identifier to uniquely identify an apparatus from a plurality of apparatus; and

outputting an audible signal from the apparatus based on the modified artificial speech signal, the audible signal being audibly different to a human listener from other audible signals output by respective ones of the plurality of apparatus for the text.

23. A non-transitory computer-readable storage medium comprising instructions that, when executed, cause a machine to at least:

convert text to an initial artificial speech signal;

generate a voice personalization signal corresponding to a subset of sounds corresponding to a subset of frequency bands in a frequency-domain representation of the initial artificial speech signal, the subset of the frequency bands selected from the initial artificial speech signal based on an identifier, the identifier defined by a string of bits, the subset of the frequency bands corresponding to the subset of sounds identified based on values of individual bits of the string of bits assigned to different ones of the frequency bands;

combine the initial artificial speech signal with the voice personalization signal to generate a modified artificial speech signal, the identifier to uniquely distinguish the machine from other machines; and

output an audible signal based on the modified artificial speech signal, the audible signal to be audibly different to a human listener than another audible signal output by a respective one of the other machines for the text.

24. The non-transitory computer-readable storage medium as defined in claim 23, wherein the instructions cause the machine to generate the voice personalization signal by:

generating the frequency-domain representation of the initial artificial speech signal;

generating a frequency-domain identifier from the identifier;

modifying the frequency-domain representation with the frequency-domain identifier to form a frequency-domain personalization signal; and

forming the voice personalization signal from the frequency-domain personalization signal.

25. The non-transitory computer-readable storage medium as defined in claim 23, wherein the identifier includes at least one of a serial number, an international mobile equipment identity (IMEI) number, a media access control (MAC) address, a device unique identifier (DUID), a processor serial number, a central processing unit (CPU) serial number, a management information base (MIB) identifier, or a Bluetooth®

address.