Voice processing apparatus and method

• US 8,315,855 B2
• Filed: 07/22/2009
• Issued: 11/20/2012
• Est. Priority Date: 07/25/2008
• Status: Active Grant

• Current Assignee: Yamaha Corporation
• Original Assignee: Yamaha Corporation
• Inventors: Yoshioka, Yasuo
• Primary Examiner(s): Vo, Huyen X.
• Application Number:
  US12/460,650
• Publication Number:
  US 20100023321A1
• Time in Patent Office: 1,217 Days
• Field of Search: 704/258, 704/260, 704/244, 704/268, 704/207, 704/266, 704/200, 704/205, 704/206, 704/209, 704/220
• US Class Current: 704/207
• CPC Class Codes: G10H 1/366 with means for modifying or correcting the external signal, e.g. pitch correction, reverberation, changing a singer's voice
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 21/003 Changing voice quality, e.g. pitch or formants

Abstract

Character extraction section extracts character amounts, pertaining to a prosody of voice, from a voice signal sequentially in a time-serial manner. Difference value calculation calculates a difference value between each of the extracted character amounts and a reference value. Processing values, corresponding to the individual character amounts, are generated in accordance with the respective difference values, and a voice processing section controls the individual character amounts of the voice signal in accordance with the processing values corresponding to the character amounts and thereby generates an output signal having a prosody changed from the prosody of the voice signal.

Claims

1. A non-transitory machine readable medium containing a program executable by a computer to perform a voice processing procedure, said voice processing procedure comprising:

a step of extracting character amounts, pertaining to a prosody of voice, from a voice signal sequentially in a time-serial manner;

a step of calculating a difference value between each of the character amounts extracted by said step of extracting sequentially in a time-serial manner and a reference value;

a step of generating processing values, corresponding to individual ones of the character amounts, in accordance with respective ones of the difference values, wherein said step of generating processing values calculates, as said processing value, a numerical value obtained by subtracting the difference value from a predetermined function value calculated using the difference value as an independent variable; and

a step of controlling the individual character amounts of the voice signal in accordance with the processing values corresponding to the character amounts to generate an output signal having a prosody changed from the prosody of the voice signal, wherein said step of controlling the individual character amounts generates the output signal by changing the individual character amounts of the voice signal by the corresponding processing values.

2. A method for changing a prosody of a voice signal, the method comprising:

a step of extracting character amounts, pertaining to a prosody of voice, from a voice signal sequentially in a time-serial manner;

a step of calculating a difference value between each of the character amounts extracted by the step of extracting sequentially in a time-serial manner and a reference value;

a step of generating processing values, corresponding to individual ones of the character amounts, in accordance with respective ones of the difference values, wherein said step of generating processing values calculates, as said processing value, a numerical value obtained by subtracting the difference value from a predetermined function value calculated using the difference value as an independent variable; and

a step of controlling the individual character amounts of the voice signal in accordance with the processing values corresponding to the character amounts to generate an output signal having a prosody changed from the prosody of the voice signal, wherein said step of controlling the individual character amounts generates the output signal by changing the individual character amounts of the voice signal by the corresponding processing values.

4. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of generating processing values calculates the processing value such that a rate of change, relative to the difference value, of the processing value increases as an absolute value of the difference value increases.

5. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of generating processing values calculates the processing value such that a rate of change, relative to the difference value, of the processing value decreases as an absolute value of the difference value increases.

6. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of generating processing values variably controls a relationship between the difference values and the processing values.

7. The method for changing a prosody of a voice signal as claimed in claim 2 which further comprises a step of setting the reference value in accordance with the character amounts extracted by said step of extracting character amounts.

8. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of extracting character amounts extracts character amounts of a plurality of types from the voice signal,

said step of calculating a difference value calculates, for each of said plurality of types, the difference value between each of the character amounts and the reference value set for the type,

said step of generating processing values generates, for each of said plurality of types, the processing values corresponding to the character amounts on the basis of the difference values, and

said step of controlling the individual character amounts controls the individual character amounts of the voice signal per each of said plurality of types.

9. The method for changing a prosody of a voice signal as claimed in claim 2 wherein the character amounts are of at least one of two types that are a volume and pitch of the voice.

10. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of generating processing values calculates a processing value corresponding to the difference value in accordance with a predetermined function.

11. The method for changing a prosody of a voice signal as claimed in claim 10 wherein said step of generating processing values changes a characteristic of the predetermined function in accordance with a parameter for controlling emphasis or depression of the prosody.

12. The method for changing a prosody of a voice signal as claimed in claim 2 which further comprises a step of generating a control parameter for controlling emphasis or depression of the prosody, and

wherein, when the prosody is to be emphasized in accordance with the parameter generated by said step of generating a control parameter, said step of generating processing values generates the processing value such that an absolute value of the processing value increases as the difference value increases, and said step of controlling the individual character amounts processes the voice signal in such a manner as to emphasize the prosody of the voice signal as the absolute value of the processing value increases, and

wherein, when the prosody is to be depressed in accordance with the parameter generated by said step of generating a control parameter, said step of generating processing values generates the processing value such that the absolute value of the processing value increases as the difference value increases, and said step of controlling the individual character amounts processes the voice signal in such a manner as to depress the prosody of the voice signal as the absolute value of the processing value increases.

13. The method for changing a prosody of a voice signal as claimed in claim 12 wherein, when the prosody is to be emphasized or depressed, the processing value is scaled in accordance with the value of the parameter.

14. The method for changing a prosody of a voice signal as claimed in claim 12 wherein said parameter is automatically generated in response to manual operation by a human operator or in accordance with a predetermined condition.

15. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of generating processing values generates the processing value in accordance with the difference value and the parameter for controlling emphasis or depression of the prosody, and

wherein, when said parameter is of a neutral value instructing neither emphasis nor depression of the prosody, said step of generating processing values does not generate the processing value irrespective of a value of the difference value, but, when said parameter is of a value instructing either emphasis or depression of the prosody, said step of generating processing values generates the processing value such that an absolute value of the processing value increases as the difference value increases.

16. The method for changing a prosody of a voice signal as claimed in claim 15 wherein, when the prosody is to be emphasized or depressed, the processing value is scaled in accordance with the value of the parameter.

17. The method for changing a prosody of a voice signal as claimed in claim 15 wherein said parameter is automatically generated in response to manual operation by a human operator or in accordance with a predetermined condition.

18. The method for changing a prosody of a voice signal as claimed in claim 2 wherein said step of controlling the individual character amounts controls the individual character amounts of the voice signal in accordance with the processing values corresponding to the character amounts to generate the output signal having the prosody of the voice signal changed to be emphasized or depressed.

3. A method for changing a prosody of a voice signal, the method comprising:

a step of extracting character amounts, pertaining to a prosody of voice, from a voice signal sequentially in a time-serial manner;

a step of calculating a difference value between each of the character amounts extracted by the step of extracting sequentially in a time-serial manner and a reference value;

a step of generating processing values, corresponding to individual ones of the character amounts, in accordance with respective ones of the difference values; and

a step of controlling the individual character amounts of the voice signal in accordance with the processing values corresponding to the character amounts to generate an output signal having a prosody changed from the prosody of the voice signal, wherein

said step of generating processing values calculates, as said processing values, numerical values obtained by subtracting the difference values from a predetermined function value calculated using the difference values as independent variables,

said step of controlling the individual character amounts generates the output signal by changing the individual character amounts of the voice signal by the corresponding processing values,

when the prosody is to be emphasized, said step of generating processing values calculates the processing values on the basis of the function value set such that an absolute value of the function value exceeds an absolute value of the difference value, and

when the prosody is to be depressed, said step of generating processing values calculates the processing value on the basis of the function value set such that the absolute value of the function value falls below the absolute value of the difference value.