JP2966954B2 - Document reading instruction device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document reading instruction device for outputting a reading position of document data by using Japanese language analysis processing.

[0002]

2. Description of the Related Art Conventionally, as a method of reading a document such as a manuscript in a fixed time, a method of reading a whole manuscript from a normal reading time required for one manuscript sheet or the like according to a human sense is used. It was presumed and this was used as an instructional material, and the reading speed was adjusted and read out. That is, the reading speed is determined based on the intuition of the person who reads. Naturally, this method requires the skill of the person who reads the text, and since there is a difference in the skill among individuals, it is expected that anyone will finish the reading correctly and accurately within the specified time. It is not possible. Furthermore, since the reading of words and accent positions in the text to be read have individual differences depending on the person who reads, the reading is not always performed with correct reading and accent.

[0003]

SUMMARY OF THE INVENTION As described above, the reading of a document to anyone can be performed within a certain period of time, and
There was no way to read aloud and accented correctly, but only to rely on the intuition and proficiency of the reader.

The present invention has been made in view of the above-described circumstances, and has as its object the purpose of making it possible for anyone to read and read a document within a certain period of time, and to read, accent, etc. It is an object of the present invention to provide a text-to-speech instruction device that can read aloud accurately.

[0005]

That is, the present invention provides a word dictionary storing word data, a document storage memory storing document data, and a reading time for the document data stored in the document storage means. A time setting unit to be set; an analyzing unit that analyzes the document data stored in the document storage unit using the word data stored in the word dictionary to obtain the reading positions and accent positions; and an analysis result of the analyzing unit. And a speed setting unit that calculates and sets a reading speed based on the reading time set by the time setting unit, and reads the document data analyzed by the analysis unit according to the reading speed set by the speed setting unit. The instruction position and the reading position of the accent position and the reading position are sequentially output by moving the display. Just read aloud according to location, can be anyone easily perform accurate reading within the set time.

[0006]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows the circuit configuration, wherein 1 is C
It is a control unit that is configured by a PU and controls the operation of the entire other circuit. Reference numeral 2 denotes a key input unit for inputting document data, various instructions, and input setting of reading time, 3 denotes a display unit constituted by, for example, a CRT, and 4 denotes a setting for storing reading time data input and set by the key input unit 2. Buffer, 5 is a document data storage unit that stores document data, 6 is a word dictionary that stores a lot of word data including word form information, reading information, accent information, etc., and 7 is stored in the word dictionary 6 A Japanese analysis unit that performs Japanese analysis of the document data read from the document data storage unit 5 based on the word data, and 8 is an analysis result buffer that holds the analysis results analyzed by the Japanese analysis unit 7 , 9 is a display data buffer for holding read-out data for reading out, 10
Is a display data creation unit for creating display data, and 11 is a reading time stored in the setting buffer 4 and a display data buffer.
This is a reading speed setting section for setting a reading speed according to the reading data stored in the reading device 9. Further, reference numeral 12 denotes a voice data rule unit that stores grammatical rules and the like for synthesizing voice, and 13 denotes an analysis data stored in the analysis result buffer 8 in accordance with rules stored in the voice data rule unit 12. An audio data generation unit that generates audio data from the result by referring to the setting buffer 4 and the analysis result buffer 8, 14 is an audio data buffer that stores the audio data generated by the audio data generation unit 13, and 15 is this audio data buffer. A speech synthesizer 16 for performing rule-based synthesis of speech based on the speech data stored in the data buffer 14 is composed of, for example, a speaker, and a speech synthesizer 15
Is a voice generating unit that converts a voice signal based on the electric signal synthesized in step 1 into an actual voice and outputs the actual voice. Next, the operation of the above embodiment will be described.

FIG. 2 shows a processing procedure for selecting and reading a document stored in the document data storage section 5 and displaying and outputting read-out data in accordance with the selected read-out time. This is mainly performed by the control unit 1.

At the beginning of the operation, the control unit 1 reads out a plurality of document data names stored in the document data storage unit 5 and outputs them on the display unit 3, and then selects any of the document data as a reading target. Or prompt for a choice. Here, the document data storage unit 5 stores document data created by a Japanese word processor and document data read by an optical character reader (OCR) in advance, and a key input unit corresponding to the display. In step 2, an instruction is given to select any document data (step S1). FIG. 3 exemplifies the contents of the document data selected in this manner, and it is assumed that the document data including the sentence "I read the book today" is selected.

Subsequently, the control unit 1 causes the display unit 3 to display and output each item of the basic reading conditions of the selected entire document data, and then prompts the user to input the setting. Corresponding to this, conditions are set in the key input section 2 (step S2). FIG. 4 exemplifies the reading conditions displayed on the display unit 3. The setting information includes, for example, reading speed, sound quality, height, strength, end setting time, presence / absence of special reading of emphasized characters,
There is a change in special reading, presence / absence of reading aloud, length of rest,..., And the like.
Is stored in the setting buffer 4.

Next, the control unit 1 sends the selected document data to the Japanese language analysis unit 7 and causes the Japanese language analysis to be performed in accordance with a large number of word data stored in the word dictionary 6 (step S3). FIG. 5 shows the data structure near the word data "I" stored in the word dictionary 6, and each word data is composed of its headline, part of speech, reading, accent, meaning and others. The Japanese analysis unit 7 analyzes the document data morphologically, syntactically and semantically based on such word data and divides it into words, and stores the analysis results in the analysis result buffer 8 by the control unit 1. Is done. FIG. 6 shows the result of analyzing the sentence "I read the book today" in the above document data.
S), number of beats, part of speech, inflected form, presence / absence of emphasis, and content. At this time, emphasized characters such as enlarged characters, inverted characters, and underlined characters are stored in the analysis result buffer 8 together with the attribute of being emphasized characters. The attribute of the emphasized character is expressed, for example, by the fact that the determined control code is immediately before the target character.

After the analysis is completed, the Japanese language analyzer 7 sends a signal to the controller 1 indicating that the analysis has been completed.
The control unit 1 initializes an internal counter (not shown) corresponding to the signal from the Japanese analysis unit 7 and
The voice data generating unit 13 reads the leading data of the analysis result stored therein, and generates the voice data based on the reference sound quality setting information while referring to the voice data rule unit 12 and the setting buffer 4 based on the analysis result. (Step S
4).

FIG. 7 shows a part of a rule stored in the voice data rule unit 12. For example, when a verb utilizing an accent (ACS) of type 0 and a five-stage inflection is an uncompleted verb. Indicates that the shape of the accent is 0 type. " The control unit 1 causes the audio data buffer 14 to temporarily hold the generated audio data.

Thereafter, the control section 1 determines whether or not the generated voice data has the attribute indicating the emphasized character as described above (step S5). Then, if there is an emphasized character, it is further determined whether or not the emphasized character is for emphasizing the reading indicated by, for example, an underline as a symbol (step S6). If it is for emphasizing the reading, additional setting of emphasized reading is performed by discriminating only a corresponding portion of the generated sound data held in the sound data buffer 14 in accordance with the reading condition of the emphasized character (step). S7). FIG. 8 shows a format of the audio data generated by the audio data generating unit 13. As shown in the figure, the audio data has a speed; a sound quality; a height; an intensity;

[0015] Here, the original text is "I've read the book today (I have been reading a book today)" is assumed to be stored in the analysis result buffer 8 are analyzed document is called. In the process of step S4, the voice data is generated based on the voice data ruler 12 and the reference sound quality setting information set in the setting buffer 4 in the process of step S4. 5;
1: 20; 20; / Kyo / phone / yomishimashita;

[0016] Audio data in the following format is generated. As shown in FIG. 8, katakana in the character string to be read out is audio data of the festival, "@" indicates an accent position, "." Indicates a pause, and "/" indicates a delimiter.

Here, if there is an attribute of the emphasized character to be emphasized and read out, and if the setting of the underlined "Kyou" is set to be emphasized and read out in step S7,
For example, 0.5; 1; 20; 20; 0.8;
1: 20; 20; Kyo; 0.5; 1; 20; 20;
Only the relevant portion is additionally set by distinguishing it according to a preset emphasis reading-out condition, such as a phone / reading device.

If it is determined in step S5 that there is no emphasized character, or if it is determined in step S6 that there is an emphasized character but that it is not for emphasized reading, the processing in step S7 is performed. Is omitted.

After the generation of the voice data and the setting process of the emphasized reading, after receiving a signal indicating the completion from the voice data generating unit 13, the control unit 1 once initializes the voice synthesizing unit 15 (step S8), and thereafter. With reference to the setting buffer 4, it is determined whether or not the reading end time has been set (step S9). If the reading end time is set, the reading speed setting unit 11 is activated to calculate the total number of beats of all the sound data held in the sound data buffer 14, and the number of rests in the document is calculated from the reading end time. , The unit time required for each beat is calculated, and the reading speed is set based on the calculated unit time (step S10).

Here, the beat is the number of the minimum units of pronunciation. For example, the word “school” is four beats,
"Society" is a three-beat word. FIG. 9 shows a part of the constituent units of a beat. The example sentence shown in FIG. 8 is a sentence of 14 beats, and the number of rests is one.

For example, according to the above-described calculation method, it is set in the setting buffer 4 that the whole sentence has 1500 beats, the number of rests is 100, and the length of each rest is 10 [msec]. In this case, in order to read this document in 5 minutes, it should be read at the rate of 0.199 seconds per beat by the calculation of ((60 × 5- (100 × 0.01)) / 1500. The information on the reading speed calculated by the reading speed setting unit 11 in this manner is set in the speech synthesis unit 15 by the control unit 1.

If it is determined in step S9 that the reading end time has not been set, the process proceeds to step S10.
The speech speed data in accordance with the preset speech condition is set in the speech synthesizer 15 without performing the above process.

Thereafter, the control unit 1 controls the audio data buffer.
The display data creation unit 10 reads the audio data held in the display data 14 and creates display data in which accents and readings are easy to understand from the audio data (step S1).
1). The created display data is sequentially stored in the display data buffer 9 by the control unit 1. When the creation of the display data is completed, the control section 1 reads the display data from the display data buffer 9 and causes the display section 3 to display and output the display data (step S12). At the same time as this display,
Reads out the audio data corresponding to the display data from the audio data buffer 14, sets the audio synthesizing unit 15 initialized in step S8, and sends out the audio data to the audio synthesizing unit 15. The voice synthesizer 15 performs the rule synthesis of the voice according to the character string in the format of the voice data, and performs the rule synthesis for the character string subsequently transmitted according to the specified speed, sound quality, height, and strength. . Then, the electric sound signal obtained by the rule synthesis is output by the sound generation unit 16 as an actual synthesized sound.

FIG. 10 shows the data displayed on the display unit 3. In the figure, the display data is
This shows an example of displaying and outputting together with the corresponding document data. The symbol “＾” in the display data indicating the reading indicates that there is an accent on the beat at the time of uttering, together with its position. .

The control unit 1 outputs the corresponding sound as described above together with the display data displayed on the display unit 3 in this manner. At this time, the control unit 1
Display the characters of the corresponding display data by means of coloring, inversion, underlining, etc. in synchronization with the output timing from 16;
It expresses where you are currently reading.

FIG. 11 shows the display screen. Here, a state in which the beat of "Kyo" of the word "today" is read aloud, and the document data displayed above is shown in FIG. While the corresponding word is expressed in reverse video, the display data created based on the audio data displayed below is expressed in reverse video, and the readout timing is displayed. Represents

Here, if the data indicating the presence or absence of the reading is set to the "absence" side by the setting buffer 4, the above-described voice data is read out by the voice synthesizing unit 15 and the voice generating unit 16. Instead, only the display output on the display unit 3 is performed. In other words, if the operator uses this function to read out a document in accordance with the display data displayed on the display unit 3, it is possible to read out the document with a correct accent at a set time. .

[0028]

As described in detail above, according to the present invention, a word dictionary storing word data, a document storage memory storing document data, and a reading time for the document data stored in the document storage means are input. A time setting section to be set;
An analysis unit that analyzes the document data stored in the document storage unit using the word data stored in the word dictionary to obtain the reading position and the accent position thereof, and sets the analysis result of the analysis unit and the time setting unit. A speed setting unit that calculates and sets a reading speed based on the read-out time,
In accordance with the reading speed set by the speed setting unit, the yoga and accent positions of the document data analyzed by the analysis unit and the reading position are sequentially instructed and output by moving display, so that the yoga and accent characters sequentially moved and displayed are displayed. It is possible to provide a text-to-speech instruction apparatus capable of making a text-to-speech text-to-speech process for anybody within a certain period of time simply by reading aloud according to a position, and making it possible to read aloud with correct reading, accent, etc. .

[Brief description of the drawings]

FIG. 1 is a block diagram showing a circuit configuration according to an embodiment of the present invention.

FIG. 2 is a flowchart showing processing contents performed by a control unit in FIG. 1;

FIG. 3 is a view exemplifying document data stored in a document data storage unit in FIG. 1;

FIG. 4 is a diagram showing read-out setting condition information stored in a setting buffer of FIG. 1;

FIG. 5 is a diagram showing a data structure of word data stored in the word dictionary of FIG. 1;

FIG. 6 is a diagram illustrating an analysis result stored in an analysis result buffer of FIG. 1;

FIG. 7 is a view showing a part of a rule stored in a voice data rule unit of FIG. 1;

FIG. 8 is a diagram showing a format of audio data generated by the audio data generation unit in FIG. 1;

FIG. 9 is a diagram showing a part of a constituent unit of a beat stored in the reading speed setting unit in FIG. 1;

FIG. 10 is a view showing a document and display data displayed on the display unit of FIG. 1;

FIG. 11 is a view showing a document and display data in a text-to-speech state displayed on the display unit of FIG. 1;

[Explanation of symbols]

1 ... Control unit, 2 ... Key input unit, 3 ... Display unit, 4 ... Setting buffer, 5 ... Document data storage unit, 6 ... Word dictionary, 7 ... Japanese analysis unit, 8 ... Analysis result buffer, 9 ... Display data Buffer, 10: display data creation unit, 11: reading speed setting unit, 12: audio data rule unit, 13: audio data generation unit, 14
... Sound data buffer, 15 ... Speech synthesis unit, 16 ... Sound generation unit.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-211661 (JP, A) JP-A-4-199420 (JP, A) JP-A-2-28971 (JP, A) JP-A-3-1990 216759 (JP, A) JP-A-1-201699 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17/21-17/28 G06F 3/16 G10L 3/00

Claims (1)

(57) [Claims] 1. A word dictionary that stores word data, a document storage unit that stores document data, a time setting unit that sets a reading time for document data stored in the document storage unit, and a word dictionary that stores the word data. Analysis means for analyzing the document data stored in the document storage means using the word data to be obtained to obtain the reading position and the accent position thereof, based on the analysis result of the analysis means and the reading time set by the time setting means. Speed setting means for calculating and setting the reading speed by reading and outputting the instruction by sequentially moving and displaying the reading position and the reading position of the document data analyzed by the analyzing means according to the reading speed set by the speed setting means. And a reading means for reading a document.