VARIABLE VOICE RATE APPARATUS AND VARIABLE VOICE RATE METHOD

• US 20080201149A1
• Filed: 04/11/2008
• Published: 08/21/2008
• Est. Priority Date: 12/27/2002
• Status: Active Grant

• Current Assignee: Hisayoshi Nagae, Kohei Momosaki, Masahide Ariu
• Original Assignee: Hisayoshi Nagae, Kohei Momosaki, Masahide Ariu
• Inventors: Momosaki, Kohei, Ariu, Masahide, NAGAE, Hisayoshi
• Granted Patent: US 7,742,920 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 2021/0135 Voice conversion or morphing

Abstract

A variable voice rate apparatus to control a reproduction rate of voice, includes a voice data generation unit configured to generate voice data from the voice, a text data generation unit configured to generate text data indicating a content of the voice data, a division information generation unit configured to generate division information used for dividing the text data into a plurality of linguistic units each of which is characterized by a linguistic form, a reproduction information generation unit configured to generate reproduction information set for each of the linguistic units, and a voice reproduction controller which controls reproduction of each of the linguistic units, based on the reproduction information and the division information.

Claims

1. A variable voice rate method of controlling a reproduction rate of voice, comprising:

generating voice data from the voice;

generating text data indicating a content of the voice data, based on the generated voice data;

generating division information used for dividing the text data into a plurality of linguistic units each of which is characterized by a linguistic form;

generating, as reproduction information concerning reproduction control of the voice for each of linguistic units, information indicating a probability with which preset ones of the linguistic units are combined in a preset order;

storing, in a first storage, the reproduction information;

selecting, from the linguistic units, combinations of linguistic units each having a probability lower than a preset value, based on the stored reproduction information and the division information; and

controlling reproduction of the voice data corresponding to the selected combinations.

2. The method according to claim 1, further comprising detecting each time for reproducing each of the linguistic units,

wherein selecting the combinations selects, from the linguistic units, combinations of linguistic units each having a probability lower than a preset value, based on the stored reproduction information and the division information, a total time for reproducing the linguistic units being not more than a preset time.

3. The method according to claim 1, wherein generating the text data includes:

preparing a speaker which outputs, to an operator, voice corresponding to the voice data; and

accepting input of text data corresponding to the voice in accordance with an operation of the operator.

4. The method according to claim 1, further comprising receiving information related to reproduction speed of the text data,

wherein generating the information generates, as the reproduction information, received information.

5. The method according to claim 1, wherein:

generating the information generates, as the reproduction information, information corresponding to each of the linguistic units; and

controlling the reproduction determines a reproduction time of each of the linguistic units, based on the stored reproduction information and the division information, and controls reproduction of each of the linguistic units in the reproduction time.

6. The method according to claim 5, wherein:

generating the division information generates division information used for dividing the text data into a plurality of the linguistic units corresponding to a word class;

generating the information generates, as the reproduction information, weighting information corresponding to each of the linguistic units; and

selecting the combinations determines the reproduction time based on the weighting information and the division information, and controls each of the linguistic units to reproduce each of the linguistic units in the determined reproduction time.

7. The method according to claim 1, wherein selecting the combinations selects to-be-reproduced linguistic units from the linguistic units, based on the stored reproduction information and the division information, the linguistic units being controlled independently to each other in reproduction.

8. The method according to claim 1, wherein:

generating the information generates a value of M (M is a positive integer) as the reproduction information; and

selecting the combinations selects to-be-reproduced every M-th linguistic unit from the linguistic units, based on the stored reproduction information and the division information, and controls the linguistic units independently to each other in reproduction.

9. The method according to claim 1, wherein:

generating the information generates a string of characters as the reproduction information; and

selecting the combinations selects, from the linguistic units, to-be-reproduced linguistic units each of which contains the string of characters, based on the stored reproduction information and the division information, and controls the linguistic units independently to each other in reproduction.

10. The method according to claim 9, wherein:

generating the information further generates a value of N (N is a positive integer) as the reproduction information; and

selecting the combinations selects, from the linguistic units, to-be-reproduced linguistic units each of which contains N or more character strings indicating the reproduction information, the linguistic units being controlled independently to each other in reproduction.

11. The method according to claim 1, comprising:

the reproduction information generation unit generates, as the reproduction information, first information indicating a string of characters, and second information indicating a level of priority and corresponding to the first information; and

selecting, from the linguistic units by priority, linguistic units each of which contains a string of characters having a higher priority level than a priority level, based on the reproduction information and the division information, and controls the linguistic units independently to each other in reproduction.

12. A variable voice rate method of controlling a reproduction rate of voice, comprising:

generating voice data from the voice;

generating text data indicating a content of the voice data, based on the generated voice data;

generating division information used for dividing the text data into a plurality of linguistic units each of which is characterized by a linguistic form;

generating, as reproduction information concerning reproduction control of the voice for each of linguistic units, information indicating a probability with which preset ones of the linguistic units are combined in a preset order;

storing, in a first storage, the reproduction information;

storing, in a second storage, the voice data and the division information;

selecting, from the linguistic units, combinations of linguistic units each having a probability lower than a preset value, based on the stored reproduction information and the stored division information; and

generating a voice signal, based on the voice data corresponding to the selected combinations.

13. The method according to claim 12, further comprising detecting each time for reproducing each of the linguistic units,

wherein selecting the combinations selects, from the linguistic units, combinations of linguistic units each having a probability lower than the preset value, based on the stored reproduction information and the stored division information, a total time for reproducing the linguistic units being not more than a preset time.