JPH05197392A - Regulation synthetic voice piece generator - Google Patents

Abstract

PURPOSE:To remarkably shorten voice piece generating time and to eliminate mis-input by confirming the output in voice of an input content at every key input. CONSTITUTION:Voice data can be generated by fetching syllable data which becomes the origin of voice piece data by a processing part 2 at every input of the syllable data, and also, the voice data generated by the processing part 2 is outputted as the voice from a speaker 5 by a voice output part 4, and after that, when accent data which becomes the origin of the voice piece data is inputted, the data is fetched in the processing part 2, and also, the voice piece data can be generated by coupling the accent data with the syllable data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rule-synthesized speech output device, etc., for use in a rule-synthesized speech unit preparation device which is used when a voice unit is frequently added.

[0002]

2. Description of the Related Art Conventionally, in a rule-synthesized speech unit creating apparatus for creating a rule-synthesized sound used in a rule-synthesized speech output device, each syllable data 101 constituting a speech unit as shown in FIG. 3 is operated by operating a keyboard. For example, when registering a phoneme "danger", the characters "ki", "ke", and "n" that compose the phoneme "danger" are added to the syllable data 10
Each of these syllable data 10 is input while sequentially inputting 1
After repeating the work of inputting the accent data 102 of the voice unit composed of 1 to create one voice unit data 103 and registering this as a dictionary in the database, by the voice unit registration work as shown in FIG. Voice information 1 to be read out by reading out each created voice unit data 103
04, or, as shown in FIG. 5, the syllable data 101 and the accent data 102, which form the voice information to be output, are collectively input to generate the voice information 104 to be output. .

[0003]

However, in the above-mentioned conventional rule-synthesized speech unit creating apparatus, the input contents cannot be confirmed for each syllable data 101.
When the accent data 102 is input after inputting all the syllable data 101, an input error of the syllable data 101 is often noticed.

Therefore, when registering a large number of voice piece data 103 or inputting a long voice piece data 103 at a time, there is a problem that it takes a long time to produce the voice piece data due to an input error.

In view of the above-mentioned circumstances, the present invention can output the input contents by voice and confirm the input each time one key input is performed, thereby significantly reducing the time required for the voice piece preparation work. At the same time, it is an object of the present invention to provide a rule-synthesized speech unit making apparatus capable of eliminating input errors.

[0006]

In order to achieve the above-mentioned object, a rule-synthesized speech unit creating apparatus according to the present invention is a rule-synthesized speech unit creating apparatus that creates speech unit data based on input data. Every time the syllable data that is the source of the syllable data is input, a syllable processing unit that captures the syllable data and creates voice data, and a syllable voice output that outputs the voice data created by this syllable processing unit as voice from a speaker And a data combining unit that, when the accent data that is the source of the speech piece data is input, captures the data and combines it with the voice data corresponding to the syllable data.

[0007]

In the above structure, every time the syllable data that is the source of the syllable data is input, the syllable processing unit takes in the syllable data to create the voice data, and the syllable voice output unit causes the syllable processing unit to output the syllable data. The created voice data is output as a voice from the speaker, and when the accent data that is the source of the voice piece data is input thereafter, this is taken in by the data combining unit and the voice data corresponding to the syllable data is obtained. The speech piece data is created by combining them.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing an embodiment of an apparatus for producing a rule-based synthesized speech unit according to the present invention.

The rule-synthesized speech unit creating apparatus shown in this figure has a key input unit 1, a processing unit 2, a storage unit 3, and a voice output unit 4.
And a speaker 5 are provided, and each time syllable data is input from the key input unit 1, a voice corresponding to the syllable data is output from the speaker 5 to allow the operator to confirm it. Then, after the input of the syllable data is completed, an accent is added to the syllable data based on the content input from the key input unit 1, a voice corresponding to the syllable data is output from the speaker 5, and this is operated. To confirm.

The key input unit 1 is provided with a standard keyboard having various keys, and when any one of the keys is operated, a key signal corresponding to the operation content is generated and supplied to the processing unit 2. To do.

The processing unit 2 includes a CPU for performing various processes,
It is provided with a ROM that stores a program and constant data that defines the operation of the CPU, a RAM that is used as a work area of the CPU, and the like, and the RAM is used based on a command output from the key input unit 1. The data output from the key input unit 1 is processed and each syllable data or accent data obtained by this process is supplied to the storage unit 3 or the syllable data or accent data designated from the storage unit 3 is read out. To generate voice piece data, to generate voice data (voice information) from the voice piece data, and to supply the voice data to the voice output unit 4.

The storage unit 3 is composed of a hard disk device, a semiconductor storage device, or the like, and when the processing unit 2 outputs syllable data or accent data, the storage unit 3 captures and stores the syllable data or the specified syllable data. The accent data is read and supplied to the processing unit 2.

The voice output unit 4 is equipped with a conversion circuit for taking in the voice data output from the processing unit 2 and converting the voice data into a voice signal, and takes in the voice data output from the processing unit 2. Then, this is converted into an audio signal and supplied to the speaker 5.

When a voice signal is output from the voice output section 4, the speaker 5 takes in the voice signal and outputs a voice.

Next, referring to the flow chart shown in FIG. 2, the voice piece data creating operation of this embodiment will be described.

First, when the keyboard of the key input section is operated to input the syllable data which forms the syllable data (step ST1), the CPU of the processing section 2 fetches and stores this, and the syllable data is stored. It is checked whether or not it is the final one (step ST2), and if this is not the final syllable data, voice data is created based on the syllable data (step ST3), and this is output to the voice output unit 4 and the speaker 5 is produced. To output a voice corresponding to the voice data (step ST4) and store the syllable data in the storage unit 3 (step ST5).

Thereafter, the CPU repeats the above-mentioned operation until the last syllable data forming the voice piece data is input (steps ST1 to ST5).

Thereafter, when the last syllable data is input and the input processing of the syllable data is completed (step ST2), if the accent data indicating the accent position is input from the key input unit 1 (step ST6), the The CPU takes in the accent data and combines it with the syllable data string stored in the storage unit 3 to create voice piece data (step ST7), and also creates voice data corresponding to the voice piece data. Is output to the voice output section, and the voice corresponding to the voice data is output from the speaker 5 (step ST8).

If there is a key input indicating the OK sign from the key input unit 1 (step ST9), the CP
If U has finished the processing on the speech piece data and there is no key input indicating the OK sign from the key input section (step ST9), the processing returns to the accent data input processing,
The above operation is repeated (steps ST6 to ST9).

As described above, in this embodiment, every time the syllable data is input, the voice data corresponding to the syllable data is created, the voice is output from the speaker 5, and the operator confirms it. When the accent data indicating the accent position is input after the last syllable data forming the syllable data is input, the accent data and the syllable data string formed by the syllable data are combined to generate the voice data. Since it is created and is output as a voice from the speaker 5 so that the operator can confirm it, the input content can be output as a voice every time one key input is performed, and this can be confirmed. It is possible to significantly reduce the creation work time and eliminate input mistakes.

[0021]

As described above, according to the present invention, 1
Each time one key is input, the input content can be output as a voice and the user can confirm this, which can significantly reduce the time required for the voice piece creation work and eliminate input mistakes.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a regularly synthesized speech unit creating device according to the present invention.

FIG. 2 is a flowchart showing an operation example of the rule-synthesized speech unit creating device shown in FIG.

FIG. 3 is a schematic diagram showing an example of a rule-synthesized voice piece creating operation in a conventionally-known rule-synthesized voice piece creating apparatus.

FIG. 4 is a schematic diagram showing an example of a voice synthesizing operation in a conventionally known rule-synthesized speech unit creating device.

FIG. 5 is a schematic diagram showing another example of the rule-synthesized speech unit creation work in the conventionally-known rule-synthesized speech unit creating apparatus.

[Explanation of symbols]

 1 key input unit 2 processing unit (syllable processing unit, data combining unit) 3 storage unit 4 voice output unit (syllable voice output unit) 5 speaker

Claims (1)

[Claims] 1. A rule-synthesized voice unit creating apparatus for creating voice unit data based on input data, imports syllable data which is a source of the voice unit data, and imports the voice data. A syllable processing unit to be created, a syllable sound output unit for outputting the sound data created by this syllable processing unit as a sound from a speaker, and when the accent data which is the source of the syllable data is input, it is captured. A rule-synthesized phonetic piece creating device, comprising: a data combining unit for combining with voice data corresponding to the syllable data.