JP2002288588A - Handwritten character input device and control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a handwritten character input device and control method capable of obtaining input feeling of words and sentences. SOLUTION: The handwritten character input device has a function of outputting the character recognition result corresponding to handwritten input by voice synthesis. The device is provided with a means 103 for recognizing one input character unit and outputting the character recognition result by voice synthesis 102, and a means 103 for recognizing one word or one sentence unit when one character unit is continuously input and outputting the recognition result by voice synthesis.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a handwritten character input device and a control method.

[0002]

2. Description of the Related Art In general, a keyboard is widely used as an input means of an information device. However, in an information device in which a keyboard is not provided or a portable device requiring a small size, a human handwriting input is required. Is input as coordinate data as it is, and a handwritten character recognition method for recognizing it as a character is used.

[0003] Algorithms for recognizing the input of handwritten characters include OCR recognition for judging characters as shadows,
On-line recognition and the like, which make decisions based on the character input procedure, have been proposed. In this online recognition, the recognition rate can be increased because the writing order information is used, as compared with the OCR recognition. However, there is a disadvantage that the recognition cannot be performed if the writing order is wrong. On the other hand, although the OCR recognition is not affected by the writing order, there is a drawback that the recognition rate is reduced due to a habit character or the like.

By the way, as an input device, a character input tablet such as a touch pad or a touch panel is used.
It has been proposed to input human input character information as coordinate data having a time axis. In this case, it is necessary for the information device to recognize a character delimiter from information that is continuously input in order to distinguish information between characters. As a method of separating characters, after a single character has been input, a specific button is pressed to notify the information device of the end of the input, or one character input is input to some extent continuously Assuming that a character has not been input for a certain period of time tl or more, a method of notifying an information device that one character input has been completed has been proposed. Here, the latter method does not require an operation such as pressing a button, so that the operability is excellent and the device is simplified.

[0005] Handwritten character input can be operated with one hand as compared with a keyboard, and has the advantage of intuitive input of characters. In addition, since it is possible to input in Chinese characters, there is an advantage that character conversion using a keyboard becomes unnecessary, and a load required for operations is reduced.
For this reason, although the input speed cannot be expected as compared with the keyboard, the usefulness of the input means under the condition that the input is performed while performing some work is recognized.

In particular, since handwritten character input is an operation to which human beings are accustomed to some extent, it becomes possible to input characters without using vision.

At the stage of handwritten character input, from the viewpoint of the operator, when one character input is completed, it is necessary to confirm how the information device has recognized the character input. Conventionally, a recognition result of one character is output by speech synthesis, and an input operation and an input result confirmation operation can be performed without using visual recognition.

[0008]

However, in the conventional configuration, when one character input is completed, the recognized character result is output only by voice synthesis. In the case of input of characters, it is not necessary to concentrate on consciousness especially during operation, but when inputting information such as words and sentences in handwritten characters without using visual information, it is very conscious of input Concentration is required.

This is because the operator has to store characters that have been previously input and then input the next character in the procedure for inputting characters.

For example, when an input result of a handwritten character is displayed on a display device and an input is performed while checking the display device, the character input at the time of the input is combined with a previously input character to form a word, Or, because it can always be grasped as a sentence, smooth character input becomes possible, but if you only check the input result of one character by voice under conditions that do not use vision, the input feeling as a word or sentence will be Cannot be obtained and character input becomes difficult.

It is an object of the present invention to provide a handwritten character input device and a control method which can solve the above-mentioned problems of the prior art, can obtain a sense of input as words and sentences, and facilitate character input. To provide.

[0012]

According to the first aspect of the present invention,
In a handwritten character input device having a function of outputting a character recognition result corresponding to handwritten character input by voice synthesis, when one character unit is input, it is recognized,
Means for outputting the character recognition result by voice synthesis;
When one character unit is continuously inputted, the unit is provided with means for recognizing one word or one sentence unit and outputting the recognition result by speech synthesis.

According to a second aspect of the present invention, in a handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by speech synthesis, when one character unit is input, the unit recognizes the input. Means for outputting the character recognition result by speech synthesis after the time t1, a buffer memory for sequentially storing the recognized characters, and a time t longer than the time t1 after the input of one character unit is completed.
Means for outputting one word or one sentence unit stored in the buffer memory by voice synthesis after two lapses.

According to a third aspect of the present invention, in the second aspect, when the information stored in the buffer memory is one character, there is provided means for omitting output by speech synthesis. I do.

According to a fourth aspect of the present invention, there is provided a handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by voice synthesis, wherein two different times t1 and t2 (t1 <t2) are set. On the other hand, when a handwritten character is input, if there is no next input for more than the time t1 after the input is detected, it is determined that one character unit has been input, and the character is recognized. Means for reading out by voice and means for continuously reading out the recognition result of the characters input so far when there is no next input for a time t2 or more after the input detection. It is assumed that.

According to a fifth aspect of the present invention, in accordance with the fourth aspect, at least a touch panel, a recognition unit, a voice synthesizing device, and a control unit are provided. The input unit recognizes the input character, reads out the character recognized by the voice synthesizer unit, compares the two times t1 and t2 with the time when there is no input, and recognizes the input character. Characteristic reading control is performed.

According to a sixth aspect of the present invention, in addition to the fifth aspect, the apparatus further comprises a display unit, wherein the recognized character is read out by the voice synthesizing unit and simultaneously displayed on the display unit. .

According to a seventh aspect of the present invention, in any one of the second to sixth aspects, the times t1 and t2 can be changed.

The invention according to claim 8 is a control method of a handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by voice synthesis, wherein two different times t1, t2 ( When t1 <t2), when a handwritten character is input, the time t1 is detected after the input is detected.
When there is no next input over the above, it is determined that one character unit has been input, the character is recognized, and the recognition result is read out by voice.
When there is no next input over two or more times, a step of continuously reading out the recognition results of the characters input so far by voice is provided.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a handwritten character input device and a control method according to the present invention will be described below with reference to the accompanying drawings.

FIG. 1 is a block diagram of a handwritten character input device. This handwritten character input device uses a handwritten character (XY
Information), a display unit 101 for displaying trajectory information of handwritten characters and a recognition result,
Voice synthesizer 102, control unit 103, recognition unit 104
And a buffer memory 105. In the case where the input operation is performed without using the visual sense, the display unit 101 is unnecessary. The touch panel 100 uses the whole as one input area.

In FIG. 2, when, for example, a character “R” is input to the touch panel 100, the character input information is sampled at a predetermined sampling rate. Is transmitted to the recognition unit 104 as input data (X, Y) at that time. A Z signal is sent as data indicating the presence or absence of an input at that time. When the Z signal is 0, it indicates that there is no input, and when the Z signal is 1, it indicates that there is an input.

FIG. 3 shows a control flow at the time of character input.

First, it is determined whether or not a Z signal is output by character input (S1). Referring to FIG. 2, since the character “RI” has two strokes, in this case, the Z signal is output twice. In S1, when the Z signal is output (Z = 1), input data (X, Y) is obtained (S1).
2) The acquisition of the input data (X, Y) is continued as long as the Z signal is output in S3. That is, the input data (X, Y) is output from the touch panel 100 to the control unit 3 during the period in which the Z signal is being output. Hereinafter, this control is controlled by the control unit 103.

Here, referring to FIG. 2, the time for recognizing the space between characters (character space) is tl. This time t
l specifies the character input speed. Note that the time tl for recognizing the space between characters is set longer than the time ta between the first and second strokes when the character “R” is input (ta <t1).

In S3, when the Z signal is not output (Z =
0), counting by a timer is started (S4). Then, it is determined whether the Z signal is output or not (S
5) If the Z signal is output (Z = 1), the timer is cleared (T = 0), and the acquisition of the input data (X, Y) is continued (S2).

In S5, when the Z signal is not output (Z =
0), timer time T and time t for recognizing the space between characters
is compared with 1 (S6). If the time T of the timer is within the time tl for recognizing the space between characters (T <t1), the process returns to S5, and it is determined whether or not the Z signal is output.
If the timer time T exceeds the time tl for recognizing the space between characters (T ≧ t1), the process shifts to character recognition (S7), shifts to speech synthesis of the recognition result (S8), and the input data (X,
Y) is cleared (S9).

That is, if the Z signal is not input again for the time tl after the input of the Z signal is confirmed,
Referring to FIG. 2, it is determined that the input of one character “R” has been completed, and the process proceeds to the recognition operation of character “R” using the input data (X, Y) up to that point. Is output from the speech synthesizer 102. In this case, if there is continued character input, the operation of input recognition is continued. The above steps S1 to S9 are almost the same as the conventional recognition process.

In this embodiment, when the character is recognized in S7, the character is sequentially stored in the buffer memory 105 at the same time.

When the input data (X, Y) is cleared in S9, it is determined whether or not the Z signal is output (S9).
10) If the Z signal is output (Z = 1), acquisition of input data (X, Y) is executed (S2). S10
When the Z signal is not output (Z = 0), the time T of the timer described above is compared with a time t2 set longer than the time tl (S11). Each time tl, t2 is
Data time tl, t2, which depends on the input speed of characters and therefore has individual differences and can be changed by individual.
It is desirable to use as.

Referring to FIG. 4, for example, characters "ri" and "n"
When "go" is input, since "ri" is two strokes, the Z signal is output in two parts, and "n" is one stroke.
Since the Z signal is output once and “go” has four strokes, the Z signal is output in four parts. After the input of the character "ri", the character "n" is input. After the input of the character "n",
The character "go" is entered. Each character "ri""n"
If the time tl has passed after the input of “go” has been completed,
In accordance with S9, each of the characters "ri", "n", and "go" is synthesized and output in units of one character for recognition confirmation.

In this embodiment, if the measured time T of the timer is within the time t2 (T <t2) in S11, the process returns to S10 to determine whether or not the Z signal is output.
In S11, if the timer time T exceeds the time t2 (T ≧ t2), in S7, the content (word “apple” in FIG. 4) stored in the buffer memory 105 is synthesized and output (S12). ), After the output of the speech synthesis, the contents are cleared (S13). When only one character (for example, “R”) is stored in the buffer memory 105, the content may be cleared without outputting the one character by speech synthesis.

In the present embodiment, if the input of the Z signal is confirmed and the character recognition is completed, and if the non-input time reaches t2, the character information stored in the buffer memory 105 up to that point is continuously spoken. As a result, the operator can obtain an input feeling as a word or a sentence, and character input becomes easy.

Further, by setting two times t1 and t2 and determining the input interval in two stages, in addition to the separation between characters, a word or one sentence is separated, and the recognition result is obtained for each unit. Can be synthesized and output.

In addition to presenting recognition results between characters by voice, automatically detecting character block units and continuously presenting the results by voice enables smooth character input. It is.

Although the present invention has been described based on one embodiment, it is apparent that the present invention is not limited to this.

For example, in the above embodiment, two times t
1 and t2 are set, and the input interval is determined in two stages. However, it is also possible to determine in three or more stages, and to confirm in a longer input unit.

[0038]

According to the present invention, since the recognition result of the handwritten character input is output by voice in units of one word or one sentence in addition to the character unit, the smooth handwritten character input can be realized even when the visual sense is not used. You.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a handwritten character input device according to the present invention.

FIG. 2 is a time chart when a character “RI” is input.

FIG. 3 is a flowchart illustrating the present embodiment.

FIG. 4 is a time chart when the word “apple” is input.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 touch panel 101 display unit 102 speech synthesis device 103 control unit 104 recognition unit 105 buffer memory

Continued on the front page (72) Inventor Masaki Tosu 5-35-2 Hakusan, Bunkyo-ku, Tokyo F-term (reference) in Clarion Co., Ltd. 5B029 BB09 CC21 EE08 5B064 AB04 BA05 DD16 FA16 5D045 AA20

Claims (8)

[Claims] 1. A handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by voice synthesis. When one character unit is input, the unit recognizes the character unit and outputs the character recognition result. Means for outputting by voice synthesis, and means for recognizing one word or one sentence unit when one character unit is successively input, and outputting the recognition result by voice synthesis. Handwritten character input device. 2. A handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by voice synthesis, wherein when one character unit is input, it is recognized, and after a lapse of time t1, a unit of character is recognized. Means for outputting the character recognition result by speech synthesis; a buffer memory for sequentially storing the recognized characters; and after the input of one character unit is completed, after a lapse of a time t2 longer than the time t1, is stored in the buffer memory. Means for outputting one word or one sentence unit obtained by speech synthesis. 3. The handwritten character input device according to claim 2, further comprising means for omitting an output by voice synthesis when the information stored in said buffer memory is one character. 4. A handwritten character input device having a function of outputting a character recognition result corresponding to handwritten character input by voice synthesis, wherein two different times t1 and t2 (t1 <t2) are set.
On the other hand, when a handwritten character is input, if there is no next input for more than time t1 after the input is detected, it is determined that one character unit has been input, the character is recognized, and the recognition result is output as a voice. Means for reading out the input, and means for continuously reading out the recognition result of the characters input so far when there is no next input for a time t2 or more after the input detection. Handwriting input device. 5. At least a touch panel, a recognition unit,
A voice synthesizing device, and a control unit. The touch panel performs input of handwritten characters, the recognition unit recognizes the input character, reads out the character recognized by the voice synthesizing unit, and reads out the character by the control unit. 5. The handwritten character input device according to claim 4, wherein the two times t1 and t2 are compared with the time when there is no input, the input character is recognized, and the character is read aloud. 6. The handwritten character input device according to claim 5, further comprising a display unit, wherein the recognized character is read out by the voice synthesizer unit and displayed on the display unit at the same time. 7. The handwritten character input device according to claim 2, wherein the times t1 and t2 can be changed. 8. A control method for a handwritten character input device having a function of outputting a character recognition result corresponding to a handwritten character input by voice synthesis, wherein two different times t1 and t2 (t1 <t2) are set.
On the other hand, when a handwritten character is input, if there is no next input for more than time t1 after the input is detected, it is determined that one character unit has been input, the character is recognized, and the recognition result is output as a voice. And, if there is no next input for a time t2 or more after the input detection, a step of continuously reading out the recognition result of the characters input so far by voice. Control method for a handwritten character input device.