Method and system for customizing voice translation of text to speech

• US 20040111271A1
• Filed: 12/10/2001
• Published: 06/10/2004
• Est. Priority Date: 12/10/2001
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: At&T Delaware Intellectual Property, Inc. (AT&T, Inc.)
• Inventors: Tischer, Steve
• Granted Patent: US 7,483,832 B2
• Time in Patent Office: Days
• US Class Current: 704/277
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser

Abstract

A method and system of customizing voice translation of a text to speech includes digitally recording speech samples of a known speaker, correlating each of the speech samples with a standardized audio representation, and organizing the recorded speech samples and correlated audio representations into a collection. The collection of speech samples correlated with audio representations is saved as a single voice file and stored in a device capable of translating the text to speech. The voice file is applied to a translation of text to speech so that the translated speech is customized according to the applied voice file.

Claims

1. A method of translation, comprising:

recording a plurality of speech samples of a speaker;

correlating the plurality of speech samples with a standardized audio representation;

organizing the plurality of speech samples and correlated audio representations into a collection;

saving the collection as a voice file; and

storing the voice file in a device capable of translating a text to speech,

wherein when the device translates the text to speech, the translated speech is customized according to the voice file.

2. The method of claim 1, wherein the speaker comprises a specific known speaker.

3. The method of claim 1, further comprising applying the voice file to a translation of the text to speech.

4. The method of claim 1, further comprising distributing the voice file to another device capable of translating the text to speech.

5. The method of claim 4, further comprising transmitting to the another device to which the voice file is distributed a subset of the voice file sufficient to cause the text to be translated to speech using the voice file.

6. The method of claim 1, wherein recording a plurality of speech samples comprises digitally recording the plurality of speech samples.

7. The method of claim 1, wherein the speech samples comprise samples of at least one of speech speed, emphasis, rhythm, pitch, pausing, and emotion of the speaker.

8. The method of claim 1, wherein the standardized audio representations comprise a phoneme.

9. The method of claim 1, wherein the standardized audio representations comprise a phone.

10. The method of claim 1, wherein the standardized audio representations comprise a biphone.

11. The method of claim 1, wherein the standardized audio representations comprise a triphone.

12. The method of claim 8, further comprising classifying the phoneme with a standardized identifier.

13. The method of claim 12, wherein the standardized identifier for the phoneme comprises a unique number and wherein the voice file comprises a sequence of the unique numbers.

14. The method of claim 1, wherein the text translated to speech comprises content accessed in a computer network.

15. The method of claim 14, wherein the content accessed in a computer network comprises an electronic mail message.

16. The method of claim 1, wherein the text translated to speech comprises text communicated through a telecommunications system.

17. A method of translation, comprising:

recording a plurality of speech samples of a plurality of speakers;

correlating the plurality of speech samples with a standardized audio representation;

organizing the plurality of speech samples and correlated audio representations for each of the plurality of speakers into a separate collection;

saving each separate collection as a voice file;

storing each voice file in a device capable of translating a text to speech; and

selecting one of the voice files,

wherein when the device translates the text to speech, the translated speech is customized according to the selected voice file.

18. The method of claim 17, wherein the plurality of speakers comprises a plurality of specific known speakers.

19. The method of claim 17, further comprising applying the selected voice file to the translation of the text to speech.

20. The method of claim 17, further comprising distributing each voice file to another device capable of translating the text to speech wherein when each device translates the text to speech, the translated speech is customized according to the selected voice file.

21. The method of claim 20, further comprising transmitting to the another device to which each voice file is distributed a subset of the selected voice file sufficient to cause the text to be translated to speech using the selected voice file.

22. The method of claim 17, wherein recording a plurality of speech samples comprises digitally recording the plurality of speech samples.

23. The method of claim 17, wherein the speech samples comprise samples of at least one of speech speed, emphasis, rhythm, pitch, pausing, and emotion of each of the plurality of known speakers.

24. The method of claim 17, wherein the standardized audio representations comprise a phoneme.

25. The method of claim 17, wherein the standardized audio representations comprise a phone.

26. The method of claim 17, wherein the standardized audio representations comprise a biphone.

27. The method of claim 17, wherein the standardized audio representations comprise a triphone.

28. The method of claim 24, further comprising classifying the phoneme with a standardized identifier.

29. The method of claim 28, wherein the standardized identifier for the phoneme comprises a unique number and wherein the voice file comprises a sequence of the unique numbers.

30. The method of claim 17, further comprising combining the plurality of voice files to create a new voice file.

31. The method of claim 17, further comprising applying the selected voice file to one portion of the text and applying another of the plurality of voice files to a different portion of the text.

32. The method of claim 17, wherein the text translated to speech comprises content accessed in a computer network.

33. The method of claim 32, wherein the content accessed in a computer network comprises an electronic mail message.

34. The method of claim 17, wherein the text translated to speech comprises text communicated through a telecommunications system.

35. A method of translation, comprising:

recording a plurality of speech samples of a plurality of specific known speakers, the speech samples comprising samples of at least one of speech speed, emphasis, rhythm, pitch, pausing, and emotion of each of the plurality of known speakers;

correlating the speech samples with a standardized audio representation, the standardized audio representations comprising a phoneme;

classifying the phoneme with a standardized identifier, the standardized identifier for the phoneme comprising a unique number;

organizing the plurality of speech samples, correlated audio representations, and standardized identifiers for each of the plurality of known speakers into a separate collection;

saving each separate collection as a voice file, the voice file further comprising a sequence of the unique numbers;

storing each voice file in a device capable of translating a text to speech; and

selecting one of the voice files,

wherein when the device translates the text to speech, the translated speech is customized according to the selected voice file.

36. A text-to-speech translation voice customization system, comprising:

a voice file comprising a plurality of recorded speech samples of a speaker, the plurality of recorded speech samples correlated with a set of standardized audio representations; and

a text-to-speech translation engine for storing the voice file and applying the voice file to a translation of a text to speech,

wherein when the translation engine translates the text to speech, the translated speech is customized according to the voice file.

37. The system of claim 36, wherein the speaker comprises a specific known speaker.

38. The system of claim 36, further comprising a distribution of the voice file to another text-to-speech translation engine.

39. The system of claim 38, further comprising a transmission to the another text-to-speech translation engine to which the voice file is distributed a subset of the voice file sufficient to cause the text to be translated to speech using the voice file.

40. The system of claim 36, wherein the plurality of recorded speech samples comprises a plurality of digitally recorded speech samples.

41. The system of claim 36, wherein the speech samples comprise samples of at least one of speech speed, emphasis, rhythm, pitch, pausing, and emotion of the speaker.

42. The system of claim 36, wherein the set of standardized audio representations comprises a plurality of phonemes.

43. The system of claim 42, wherein the set of standardized audio representations comprises a plurality of phones.

44. The system of claim 36, wherein the set of standardized audio representations comprises a plurality of biphones.

45. The system of claim 36, wherein the set of standardized audio representations comprises a plurality of triphones.

46. The system of claim 42, further comprising a classification of each of the plurality of phonemes with a standardized identifier.

47. The system of claim 46, wherein the standardized identifier for each of the plurality of phonemes comprises a unique number and wherein the voice file comprises a sequence of the unique numbers.

48. The system of claim 36, wherein the text translated to speech comprises content accessed in a computer network.

49. The system of claim 48, wherein the content accessed in a computer network comprises an electronic mail message.

50. The system of claim 36, wherein the text translated to speech comprises text communicated through a telecommunications system.

51. A text-to-speech translation voice customization system, comprising:

a plurality of voice files, each voice file comprising a plurality of recorded speech samples of a different speaker, the plurality of recorded speech samples correlated with a set of standardized audio representations; and

a text-to-speech translation engine for storing the plurality of voice files and applying the plurality of voice files to a translation of a text to speech,

wherein when one of the voice files is selected, the translation engine translates the text to speech according to the selected voice file.

52. The system of claim 51, wherein each different speaker comprises a specific known speaker.

53. The system of claim 51, wherein the plurality of recorded speech samples comprises a plurality of digitally recorded speech samples.

54. The system of claim 51, wherein the speech samples comprise samples of at least one of speech speed, emphasis, rhythm, pitch, pausing, and emotion of each of the plurality of speakers.

55. The system of claim 51, further comprising a distribution of the plurality of voice files to another text-to-speech translation engine wherein when each translation engine translates the text to speech, the translated speech is customized according to the selected voice file.

56. The system of claim 55, further comprising a transmission to the another text-to-speech translation engine to which the plurality of voice files is distributed a subset of the selected voice file sufficient to cause the text to be translated to speech using the selected voice file.

57. The system of claim 51, wherein the set of standardized audio representations comprises a plurality of phonemes.

58. The system of claim 51, wherein the set of standardized audio representations comprises a plurality of phones.

59. The system of claim 51, wherein the set of standardized audio representations comprises a plurality of biphones.

60. The system of claim 51, wherein the set of standardized audio representations comprises a plurality of triphones.

61. The system of claim 57, further comprising a classification of each of the plurality of phonemes with a standardized identifier.

62. The system of claim 61, wherein the standardized identifier for each of the plurality of phonemes comprises a unique number and wherein each voice file comprises a sequence of the unique numbers.

63. The system of claim 51, further comprising a combination of the plurality of voice files to create a new voice file.

64. The system of claim 51, wherein the text translated to speech comprises content accessed in a computer network.

65. The system of claim 64, wherein the content accessed in a computer network comprises an electronic mail message.

66. The system of claim 51, wherein the text translated to speech comprises text communicated through a telecommunications system.