JPS61196360A - Word processor - Google Patents

Abstract

PURPOSE:To save the labor of the reinputting of the operator by making part of the inputted sentences into the word for recognizing the voice defined by the user. CONSTITUTION:While the voice input processing is executed in accordance with the program in a memory 32, a word registering key is pushed down, and then, a CPU 31 matches a cursor to the characters at the head and the end of part of the inputted sentences, stores the start end address of a displaying memory 33 to a memory and the inverting displaying is executed. Next, the extracting character column is uttered, the voice is inputted and the column is registered as the standard pattern of a word number (m). Next, since the operator executes the registered voice input and then, the word number (m) is sent at a voice input part 2, the CPU 31 refers to the x+m address of the memory 32, reads the contents of the address from a B address in which the definition character column is included as an address, and executes the input processing as an input character column.

Description

DETAILED DESCRIPTION OF THE INVENTION 1.1 The present invention relates to a word processor, and more particularly to an audio word processor that can perform text creation processing not only by input from a keyboard but also by voice input by an operator.

Prior Art Conventionally, in this type of voice word processor, the following processing is required to enable the operator to input a character string defined by himself or herself by voice.

In other words, first, even if text is being input, the process is temporarily stopped, another program is called for voice registration, and the desired character string is input again, and then the operator inputs the character string using the keyboard. This method involves raising the cursor position while generating the sound, and registering the voice pattern.

In the conventional voice word processors described above, even if a frequently used character string occurs during input by an operator, the voice pattern cannot be registered immediately on the spot, so it is necessary to register it after input processing or after interrupting input. there were.

However, when registering, the operator may forget what he or she wants to register, and if the input process is interrupted, the processing speed decreases.

Furthermore, the character string that the operator wants to define must be input again during execution of the registration program, which is a burden on the operator.

Purpose of the Invention The present invention has been made in order to eliminate the drawbacks of the above-mentioned conventional methods, and its purpose is to provide a system with good operability by making it possible to easily register the sound patterns of words to be registered while writing sentences. Our goal is to provide a voice word processor that is easy to use.

l1211 According to the present invention, a registration instructing means for instructing word registration of a character string of a part of an input sentence during text input using a keyboard, and a means for instructing an operator to generate a sound of the character string; i! the voice pattern uttered in response to this instruction! The device includes a voice registration means for recognizing and registering the voice pattern, and a means for storing a correspondence table showing the correspondence between the registered voice pattern and the character string, and a means for storing a correspondence table indicating the correspondence between the registered voice pattern and the character string. The result is a word processor that becomes possible.

Fire 3 Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing the overall configuration of an embodiment of the present invention. In a voice word processor, the voice input through the microphone 1 is usually compared with a pre-registered voice pattern (called a standard pattern) by the voice input section 2, and based on the comparison result, the voice input is similar to the input voice. The top n word numbers are sent to the main body 3.

CPU in main body 3 <data processing bag @) 31,
The sent word number is converted into one character code or several character code strings and stored in the input canister in the CPU.

This character code is also written into the screen display memory 33 and displayed on the display device 4.

The input character string is stored in the memory 32 within the main body, but can also be output and stored in the external storage device 6. The input character string can also be printed using the printing device @7. Furthermore, the keyboard 5 is used as an aid for voice input and voice registration. Data is exchanged between the keyboard 5, external storage device (floppy disk) 6, and printer 7 and the main body 3 via an Ilo (input/output) port 34.

The voice input section 2 is configured as shown in FIG. N.W.A.

The sounds uttered by the speaker are then compared to a standard pattern,
A code indicating a candidate with a high degree of similarity is sent to the opposite party.

FIG. 3 shows a diagram of a display device of an audio word processor, and (A) shows the input of "fifth generation computer type arithmetic device". The operator noticed that the word "computer" often appeared in sentences and converted it into spoken words (letters that can be entered at once by voice).
When the operator wants to register it as ``J'', the operator operates the cursor movement key to move the cursor to ``Computer'' and presses the ``Register'' key. Furthermore, the operator moves the cursor to "Yu" and presses the "Execute" key. Here the figure (
As shown in B), all the letters "Computer" are displayed in reverse video, and the guide message "Please speak" is displayed.The operator then speaks the word "Computer" and the sound "Computer" is stored in the voice pattern memory 23. The pattern is registered as a standard pattern, the guide message disappears, and the registration of the a voice word ends.

The next time the operator needs to input the word "computer," he or she does not have to enter the word "computer" one character at a time.
If you say the word "computer" all at once, you can input it faster.

Note that the voice word processor is configured so that input can be made using either the keyboard 5 or the voice input section 2. This is done so that you can input using the keys when the surrounding noise becomes loud or your throat gets tired, or to assist with voice registration as mentioned earlier.

Next, a program for processing voice input will be explained. The processing at CPLJ31 in the case of registering the word "computer" mentioned earlier will be described.

The CPU 31 performs voice input processing according to a program in the memory 32. When a key is pressed as part of the input processing program, the program routine shown in FIG. 4 is executed.

First, in step 51, it is determined whether or not the word registration key has been pressed. If No, no processing is performed and the process proceeds to the next step. If YES, the address of the display memory 33 where the cursor is located (in this case, the address of "K") is stored in the memory 32. Next, it waits for whether the execution key is pressed or the cancel key is pressed, and when the execution key is pressed, the address of the Kanle position (here, the address of "Yu") is stored in the memory 32 as the end address as before. Store (step 52).

Next, the CPU displays the characters from the start address to the end address in reverse video (step 53), reads out the character code string (here, "computer") highlighted from the display memory 33, and stores it in the memory 32. (Step 54).

Next, input the next voice pattern to the voice input section 2.
A command is sent to instruct it to be registered as a standard pattern with IIg number m (step 55). Here, the word number m is a number added in the order of registration. It is assumed that the word numbers and definition character strings correspond as shown in FIG.

In the memory 32, as shown in FIG. 6, a value B indicating the address containing the definition character string for word number m is stored at address X0m (x is a constant value), and from this address B, teeth"
The character string "Computer" is stored. The correspondence table is stored in the memory 32 in this format.

Next, the CPU 31 displays a guide message "Please speak" on the screen (step 56). When the voice input section 2 detects the operator's utterance, it stores the voice pattern in the voice pattern memory 23 as a standard pattern of word number m and sends a notification of completion of registration to the main body 3. When the CPU 31 receives this notification, it returns to the previous input processing (step 57).

Next, in the case of voice input, when the operator utters "computer", the voice input section 2 sends out the word number m, so the CPU 31 refers to address x+m in the memory 32 and writes the contents of this address to the address. , read from address B, read up to the end code, and know that this is a "computer". Input processing is performed using this word "computer" as an input character string.

As described in detail, according to the present invention, a part of an input sentence can be converted into a user-defined voice during input processing! l! By using words for II, there is no need to completely complete the input process, and already input sentences can be used as words defined by the user, so the operator does not have to re-enter the words, making it easier to use. Become.

Furthermore, the program required to register voice patterns and obtain registered words can be configured as part of the input processing program that is currently inputting sentences, or can be stored in memory at the same time and used as part of the input program. If the configuration is such that it is called and operated, it is possible to return the processing to the part in the middle of the input without completely terminating the input processing.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a block diagram showing the configuration of the audio input in Fig. 1, and Fig. 3 (A
＞, (B) are diagrams each showing an aspect of the display screen at the time of word registration, Figure 4 is a diagram showing the operational flow of the embodiment of the present invention, and Figure 5 is Single II! FIG. 6 is a diagram showing a correspondence table between 1NO0 and definition character strings, and FIG. 6 is a diagram showing the correspondence between word numbers and definition character strings in memory. Explanation of symbols of main parts 1...Microphone 2...Audio input unit 4...Display device 5...Keyboard 22...Data processing Section 23...Pattern memory 31...CPU 32...Memory

Claims (1)

[Claims] A voice word processor capable of performing sentence creation processing not only by input from a keyboard but also by voice input by an operator, comprising registration instructing means for instructing to register a character string of a part of an input sentence as a word during input of a sentence by a keyboard; means for instructing an operator to generate a voice of a character string; voice registration means for recognizing and registering a voice pattern generated in response to the instruction; and the registered voice pattern and the character string. and means for storing a correspondence table indicating a correspondence relationship between the word processor and the word processor, the word processor being capable of inputting a registered word based on the generation of the character string.