Voice processing apparatus for varying the speed with which a voice signal is reproduced

• US 5,189,702 A
• Filed: 10/02/1991
• Issued: 02/23/1993
• Est. Priority Date: 02/16/1987
• Status: Expired due to Term

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Tamura, Junichi, Sakurai, Atsushi
• Primary Examiner(s): Kemeny, Emanuel S.
• Assistant Examiner(s): Doerrler, Michelle
• Application Number:
  US07/770,136
• Time in Patent Office: 510 Days
• Field of Search: 381/29-40, 381/51-53, 395/2
• US Class Current: 704/266
• CPC Class Codes: G10L 21/04 Time compression or expansion

Abstract

A voice processing apparatus capable of varying the speed of speech, in which a voice of a predetermined duration is represented by feature parameters and propriety information indicating whether a change in the speech speed is permitted or not. During voice synthesis, the speech speed is varied by skipping or repeating only the feature parameters for which the variation in speech speed is permitted by the associated propriety information.

Claims

1. A voice processing apparatus for increasing the speed of synthesized speech synthesized by a voice synthesizer, comprising:

memory means for storing a plurality of sets of feature parameters, and for storing information for enabling speech speed control in such a manner as not to skip at least each set of the feature parameters in accordance with whether each set of feature parameters represents the timing of a non-stable portion of generated speech, the information being established based on the duration of speech generated using at least one feature parameter; and

speed control means for, during voice synthesis in which a voice signal is synthesized by the voice synthesizer, skipping the sets of the feature parameters, for which the speed control is enabled by the information.

2. A voice processing apparatus according to claim 1, wherein the information for enabling or disabling speech speed control is established irrespective of whether the synthesized speech is voiced or unvoiced.

3. A voice processing apparatus for decreasing the speed of synthesized speech synthesized by a voice synthesizer, comprising:

memory means for storing a plurality of sets of feature parameters, an for storing information for enabling speech speed control in such manner as not to repeat at least each set of the feature parameters in accordance with whether each set of feature parameters represents the timing of a non-stable portion of generated speech, the information being established based on the duration of speech generated using at least one feature parameter; and

speed control means for, during voice synthesis in which a voice signal is synthesized by the voice synthesizer, repeating the sets of the feature parameters, for which the speed control is enabled by the information.

4. A voice processing apparatus according to claim 3, wherein the information for enabling or disabling speech speed control is established irrespective of whether the synthesized speech is voiced or unvoiced.

5. A voice processing apparatus comprising:

memory means for storing a plurality of sets of feature parameters, used for voice synthesis by a voice synthesizer and for storing multi-value information for enabling or disabling speech speed control for at least each set of the feature parameters;

threshold value setting means for setting a threshold value in response to the speed with which a voice signal is to be synthesized by the voice synthesizer; and

speed control means for, during voice synthesis in which the voice signal is synthesized by the voice synthesizer, skipping or repeating the sets of feature parameters whose corresponding multi-value information are smaller than the threshold value.

6. A voice processing apparatus according to claim 5, wherein said memory means stores a maximum multi-value information corresponding to the feature parameters representing the explosion of plosive consonants, and multi-value information decreasing in value corresponding to the succeeding sets of feature parameters succeeding the sets of feature parameters representing the explosion of plosive consonants.

7. A voice processing apparatus according to claim 5, wherein said speed control means does not repeat the sets of feature parameters unconditionally if the corresponding multi-value information has a predetermined sign.

8. A voice processing apparatus according to claim 5, wherein said threshold value setting means sets a higher threshold value as the speed with which the voice signal is to be synthesized becomes faster or slower than a standard speed.

9. A voice processing method for increasing the speed of synthesized speech synthesized by a voice synthesizer, comprising the steps of:

storing a plurality of sets of feature parameters, and storing information for disabling speech speed control in such a manner that the set of feature parameters are not skipped for at least each set of the feature parameters in accordance with whether each set of feature parameters represents the timing of a non-stable portion of generated speech, the information being established based on the duration of speech generated using at least one feature parameter; and

skipping the sets of the feature parameters, for which the speed control is enabled by the information, during voice synthesis in which a voice signal is synthesized by the voice synthesizer.

10. A voice processing method according to claim 9, wherein said storing step comprises the step of storing the information for enabling speech speed control irrespective of whether the synthesized speech is voiced or unvoiced.

11. A voice processing method for decreasing the speed of synthesized speech synthesized by a voice synthesizer, comprising the steps of:

storing a plurality of sets of feature parameters, and storing information for enabling speech speed control in such a manner that the set of feature parameters are not repeated for at least each set of feature parameters in accordance with whether each set of feature parameters represents the timing of a non-stable portion of generated speech, the information being established based on the duration of speech generated using at least one feature parameter; and

repeating the sets of the feature parameters, for which the speed control is enabled by the information, during voice synthesis in which a voice signal is synthesized by the voice synthesizer.

12. A voice processing method according to claim 11, wherein said storing step comprises the step of storing the information for enabling speech speed control irrespective of whether the synthesized speech is voiced or unvoiced.

13. A voice processing method comprising the steps of:

storing a plurality of sets of feature parameters used for voice synthesis by a voice synthesizer, and storing multi-value information for enabling or disabling speech speed control for at least each set of the feature parameters;

setting a threshold value in response to the speed with which a voice signal is to be synthesized by the voice synthesizer; and

skipping or repeating the sets of feature parameters whose corresponding multi-value information are smaller than the threshold value, during voice synthesis in which the voice signal is synthesized by the voice synthesizer.

14. A voice processing method according to claim 13, wherein said storing step further comprises the steps of storing maximum multi-value information corresponding to the feature parameters representing the explosion of plosive consonants and storing multi-value information decreasing in value corresponding to the succeeding sets of feature parameters succeeding the sets of feature parameters representing the explosion of plosive consonants.

15. A voice processing method according to claim 13, wherein said skipping or repeating step comprises the step of not repeating unconditionally the sets of feature parameters if the corresponding multi-value information has a predetermined sign in said skipping or repeating step.

16. A voice processing method according to claim 13, wherein said setting step comprises the step of increasing the threshold value that is set in said setting step as the speed with which the voice signal is to be synthesized becomes faster or slower than a standard speed.

17. A voice processing method according to claim 13, wherein said storing step comprises the step of storing the information for enabling or disabling speech speed control irrespective of whether the synthesized speech is voiced or unvoiced.