JPS6135573B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a kanji processing method, and more particularly, to a kanji processing method suitable for application to a kanji processing device that performs display or recording and audio output.

In recent years, so-called kanji processing systems (kanji processing devices), which incorporate kanji display devices and kanji printers into electronic computer systems, have been put into practical use, making it easier to transmit information from computers to users. We are trying to Furthermore, in order to provide even more convenience, systems have been developed that transmit information from a computer to a user by voice.

The above-mentioned kanji display devices and kanji printers have the advantage that they can output kanji characters at a high speed and can record and save them, but they have the disadvantage that the user feels tired because they have to visually recognize the kanji characters. On the other hand, voice output devices place fewer constraints on users and can transmit information to many users simultaneously with one device, but their output speed is low and recording storage is difficult on their own. It has the disadvantage that it cannot be used.

Therefore, in order to provide even more convenience, a kanji processing device is provided that is equipped with both the kanji display device and kanji printer, and an audio output, so that the same message can be simultaneously conveyed to the user through kanji display and audio output. It is desired that the information be transmitted in the following manner, or in an appropriate manner.

However, conventional audio output devices convert kana into speech, and unlike kana characters, there are multiple readings for a single character type, so it is difficult to encode sentences containing kanji. When handling data as computer data, conventional codes that only identify character types cannot be used. Therefore, after converting kanji to kana, each sentence is entered into a computer separately from the data for printing and display.
It is necessary to store it in this memory. Therefore, when trying to output audio based on data that has been conventionally stored as kanji codes, there is a problem that it takes a lot of effort and the amount of data doubles.

Thus, the present invention aims to solve the above problem by utilizing codes for printing and display as codes for speech.The present invention aims to solve the above problem by using an image dictionary memory that stores images of kanji characters, Applicable〓〓〓〓
An image output means for outputting an image selectively read out from an image dictionary memory according to a kanji code; an audio dictionary memory storing audio information of kanji; and an image outputting means for outputting an image selectively read from the audio dictionary memory according to a kanji code. In the kanji processing system, the kanji processing system includes a voice output means for converting the voice information into voice, the kanji code is composed of a character type identification unit and a reading recognition unit that identifies the reading of the same character type, and the character type identification unit This is achieved by having the character type identifying part and the pronunciation identifying part select the audio information to be read from the audio dictionary memory.

That is, in the present invention, the reading identification section need only indicate which reading is to be selected from at most several predetermined readings for each character type, so that the amount of information in one byte is reduced. This is extremely advantageous compared to the conventional method of converting kanji to kana, which requires an average of several bytes of code per character.

Embodiments of the present invention will be described in detail below with reference to the drawings.

1 is a diagram illustrating a schematic configuration of a kanji output device capable of implementing the present invention, FIG. 2 is a diagram illustrating an example of a detailed configuration of the voice dictionary section 6 shown in FIG. This figure is a diagram conceptually showing an example of the operation of the speech dictionary section 6 shown in FIG. 2, and FIG. 4 is a diagram illustrating a Kanji code corresponding to the example of operation shown in FIG. 3.

In FIG. 1, 1 executes data processing according to the program, thereby selecting the kanji codes C ₁ and C ₂ and the control signal etc. for selecting their output format.
The information processing device 2 that outputs CNTL is a control circuit that controls each part of the device based on the control signal CNTL from the information processing device 1. For example, it determines the output format of kanji, that is, whether to display images, output audio, or both. 3 is _a selection circuit that distributes kanji codes based on _this control, and 4 is a kanji code C given via selection circuit 3. ₁ , a buffer memory for temporarily storing _C2 and transmitting it in accordance with the voice output speed; 5 stores a character type identification part _C1 , which is a part of the kanji code given via the selection circuit 3, for one screen. Refresh memory 6 is the kanji code C ₁ , C ₂ sent from buffer memory 4
7 is a voice dictionary section that converts the kanji codes C ₁ and C ₂ into a corresponding series of phoneme parameters, 7 is a voice output section that produces sounds based on this series of phoneme parameters, and 8 is a memory that stores character patterns. 9 is a character pattern generator that generates a character pattern signal corresponding to the kanji code periodically sent out by the refresh memory 5, and 9 is a display that displays characters on the screen using the character pattern signal generated from the character pattern generator. Department.

The operation when displaying kanji characters as images is similar to that of conventional kanji devices. That is, the character type identification part C 1 of the kanji codes C ₁ and C ₂ given from the information processing device ₁ is stored in the refresh memory 5 via the selection circuit 3, and the characters sequentially sent out from the refresh memory 5 are stored in the refresh memory 5 via the selection circuit 3. The type identification section _C1 is converted into a character pattern signal by a character pattern generator 8 to form characters on the screen. on the other hand,
When performing audio output, all of the kanji codes C ₁ and C ₂ given from the information processing device 1, in other words, the character type identification section C ₁ and the reading identification section C ₂ , are sent to the buffer memory 4 via the selection circuit 3. is stored in Then, the kanji codes C ₁ and C _{2 are sequentially sent from the buffer memory 4 at a speed that humans can understand, and as a result, a series of phoneme parameters are obtained from the speech dictionary section 6 as described later, and the speech output section 7 outputs the kanji codes C 1 and C 2} . occur.

Now, in the present invention, an operation equivalent to the operation of converting kanji into kana for audio use is automatically performed in the audio dictionary section. and,
This operation is performed by combining the character type identification unit C ₁ and the reading identification unit C ₂ for identifying readings for the same character type, so the amount of information in the reading identification unit C ₂ can be reduced. This operation will be specifically explained below based on the embodiment shown in FIGS. 2 to 4.

In FIG. 2, 6 ₆ is a buffer register into which the character type identification section C ₁ is input, 6 ₇ is a buffer register into which the reading identification section C ₂ is input, and G ₁ is the character indicated by the character type identification section C ₁ . A first address conversion circuit ₆₂ having a memory storing character type address information corresponding to the type adds the character type address information from the first address conversion circuit ₆₁ and the reading identification section _C2 . The adder ₆₃ corresponds to the reading address information represented by the output of the adder ₆₃ .
A second address conversion circuit records the first address in the dictionary memory ₆₄ of data consisting of a kana sound string representing the reading of a kanji, and the number of kana sounds included in the reading data of one character; ₆₄ is the above kana sound string; Dictionary memory 6 records reading data consisting of ₅ is dictionary memory 6
A phoneme generation circuit ₆₈ generates a group of phoneme parameters corresponding to each kana sound sequentially read out from the dictionary memory 64, and a sound number control circuit ₆₈ controls the readout of reading data from the dictionary memory ₆₄ according to the number of kana sounds.

Next, the operation will be explained. The kanji to be output now is ``Ko'', and as shown in Figure 4, for example, the kanji ``Ko'' is composed of a character type identification part C ₁ which is "0002" and a reading identification part C ₂ which is "02". It is assumed that As is well known, the kanji ``Ko'' has four readings: ``Ooyake,''``Kimi,''``Kou,'' and ``Ku.'' In this example, when specifying the first reading ``Ooyake,''00" is set in the reading identification section C ₂ of the kanji code, and when specifying the second reading, "01" is set in the same identification section C _2. Similarly, when specifying "Kou", "02" is set. When specifying "ku", "03" is set in the respective pronunciation identification section _C2 . Note that it is also possible to specify the reading of kanji other than ``Ko'' in the same way.

In this way, for example, the character type identification section C ₁ can specify the kanji ``Ko'', and the reading identification section C <sub>2</sub> can specify ``Kou'', but then the character specified according to Figure 3 can be specified. The operation of selecting sounds according to the species and how to read them will be explained.

In the figure, "2", which is the character type identification part stored in the buffer register ⁶⁶ , represents address #2 of the first address conversion circuit ₆₁ , and therefore, from the same address conversion circuit ₆₁ to Contents of #2 “3”
is read out. What this “3” means is:
In the second address conversion circuit ₆₃ , this is the first address among addresses #3 to #6 in which reading-specific information corresponding to the Kanji character "Ko" is stored. Therefore, by adding the content "2" of the pronunciation identifying section _C2 to this first address "3" using an adder, a "public" information storage address #5 having the corresponding pronunciation is obtained.

Then, when the contents of the second address conversion circuit ₆₃ are read out according to this storage address #5, "8" and "2" are obtained. Here, "8" is the pronunciation "Kou" in the dictionary memory ₆₄ that stores the pronunciation data.
This is the storage address of the data corresponding to , and similarly, "2" is the length of the kana string of the reading "Kou".
The same is true for pronunciations other than "Kou", such as "Ooyake", and the address storage section _63A of the second conversion circuit ₆₃ has a dictionary memory for the first kana sound "O" of "Ooyake". The content indicating the storage address #2 in ₆₄ is stored, and the length "4" of the kana string of the pronunciation is stored in the length storage section _63B .

Next, among the contents "8" and "2" read out from the second address conversion circuit ₆₃ in this way, "8" indicating the data storage address in the dictionary memory is used for reading out the dictionary memory ₆₄ . As a result, the phoneme address information "09" corresponding to the kana sound "ko" is read from the same address #8, and then the phoneme generation circuit ₆₅
In , the phoneme parameter "ko" is selectively generated by the homophone address information "09". When the generation of one kana sound is completed, the content "02" corresponding to the next kana sound "U" in the reading data "Kou" is read out from the dictionary memory ₆₄ , and the same generation is performed. The end of the reading data for one character is determined by the content read from the length storage section _63B , and in this example, it is specified that the generation of one character is terminated by the generation of two kana sounds. ing.

In this way, one character is generated, and by repeating this operation, one sentence or a series of sentences is outputted as audio.

In the above embodiment, utterance is performed by associating phoneme parameters for each kana sound, but instead of this, phoneme parameters for each word or phrase are generated all at once, and voice output is thereby generated. You may. Furthermore, the present invention can be similarly applied to different ways of reading, including changing only the volume, speed, and pitch of the generated sounds. Furthermore, the present invention can be applied to a terminal device that receives Kanji codes via a communication line.

As described in detail above, according to the present invention, it is possible to output both the image and audio of Kanji characters using a code with a small amount of information.

Furthermore, since existing Kanji codes for image output can be used, data can be easily changed when a new audio output device is installed, and the capacity of a file for storing data can be reduced. In addition. This〓〓〓〓
When changing data, the pronunciation identification section to be added may be directly input in the form of a code, or the pronunciation may be input in kana and internally converted into a code.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of the kanji processing device according to the present invention, FIG. 2 is a diagram illustrating the detailed configuration of the speech dictionary section shown in FIG. 2, FIG. 3 is an operational explanatory diagram, and FIG. 4 is a kanji code It is a figure showing one example of composition of. 6...Speech dictionary section, 7...Speech output section, 8...
Character pattern generator, 9...display section, _C1 ...character type identification section, _C2 ...pronunciation identification section. 〓〓〓〓

Claims (1)

[Claims] 1. An image dictionary memory that stores images of kanji, an image output means that outputs images selectively read out from the image dictionary memory according to kanji codes, an audio dictionary memory that stores audio information of kanji, and an image dictionary memory that stores images of kanji. In a kanji processing system equipped with a voice output means that converts voice information selectively read out from a voice dictionary memory according to a kanji code into voice, the kanji code is used as a character type identification unit to identify the reading of the same character type. The character type identification unit selects an image to be read from the image dictionary memory, and the character type identification unit and the reading identification unit select audio information to be read from the audio dictionary memory. A kanji processing method that is characterized by making it perform.