JPH07334075A - Presentation supporting device - Google Patents

Abstract

PURPOSE:To confirm the presentation contents and method by a user as soon as the original of the presentation is completed and to evaluate the training result of the presentation by analyzing the generated document, converting the analyzed document into a presentation expression and reading the presentation through the use of voice synthesis. CONSTITUTION:The generated document is analyzed in a document analysis means 102 using a keyboard input means 100 or a mouse input means 101, a voice input means 104 and a voice recognition means 105. The attribute value as a result of the analysis conducted by the means 102 is outputted to a presentation expression converting means 103. Moreover, a presentation condition specifying means 108 outputs the conditions related to the conversion method of the presentation expression display inputted by the means 100 or 101 to the means 103. A parameter setting means 107 outputs the parameters added to the document presentation expression display which is the output of the means 103 to a voice synthesis means 109.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document preparation application such as a word processor, and in particular, it assists in preparation of a manuscript for presentation, confirmation of a presentation, and training, and is suitable for presentation support suitable for improving presentation skills of users. Regarding the device.

[0002]

2. Description of the Related Art In a document preparation application, when preparing a manuscript for a presentation, in addition to the accuracy of the input document, a function for confirming the time required for the presentation and the comprehension of the manuscript as a presentation is required. It To that end, the system
The speed at which the manuscript is read can be changed freely,
At the same time, it is necessary to read a manuscript containing objects such as figures. However, in the conventional document creation application, the reading function reads the manuscript at a constant speed,
At that time, the manuscript to be read was only a text or a text such as an item in the table. Also, on the other hand,
In order to train the user's presentation, the ability to evaluate the user's presentation is required. In the conventional document creation application and presentation tool, there is no function for evaluating the utterance method and utterance.

[0003]

SUMMARY OF THE INVENTION An object of the present invention is to provide a presentation support function in a document creation application for confirming a presentation and training a presentation, which cannot be dealt with by the conventional techniques.

[0004]

[Means for Solving the Problems] Among the above problems, in order to solve the problem at the time of confirming the presentation, a means for analyzing the created document and a means for converting the analyzed document into a presentation format expression. And, it was equipped with a means to read out the presentation style expression by voice synthesis. Further, among the above problems, in order to solve the problem at the time of presentation evaluation, a means for analyzing the voice of the user during the presentation practice, and a means for evaluating acoustic characteristics such as speech speed, speech time, and intonation. Equipped with.

[0005]

[Function] By analyzing the created document and reading the analyzed document converted into the presentation format by voice synthesis, the user can confirm the content and method of the presentation at the same time as creating the manuscript for the presentation. Will be possible. In addition, by analyzing the voice of the user during the practice of the presentation and evaluating the acoustic characteristics such as the speech speed, the speech time, and the intonation, it is possible to evaluate the practice of the presentation of the user.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing an embodiment of a presentation support device according to the present invention. In FIG.
The means used for the presentation confirmation function are keyboard input means 100, mouse input means 101, document analysis means 102, presentation format conversion means 103, speech input means 104, speech recognition means 105, parameter extraction means 10
6, parameter setting means 107, presentation condition designating means 10
8, a voice synthesizing means 109, a voice output means 110, a screen display means 111, a screen output means 112, an information synchronization output means 113, and a voice reproduction means 114.

The keyboard input means 100 is a keyboard used for creating a document. Mouse input means 10
A mouse 1 is used for creating a document and setting parameters. The document analysis means 102 is means for analyzing a document created using the keyboard input means 100 or the mouse input means 101. Presentation format conversion means 10
Reference numeral 3 is a means for converting a character string or a figure in a document into a presentation-style expression. The voice input means 104 is a tool such as a microphone used for inputting voice. The voice recognition unit 105 is a unit that analyzes the input voice and outputs a corresponding word string and other analysis results.
The parameter extraction unit 106 is a unit that extracts the acoustic feature parameter of the input voice. The parameter setting unit 107 is a unit that adds the acoustic feature parameter of the input voice to the character representation in the presentation format, calculates the presentation time and speed, and outputs the result. The voice synthesizer 109
It is means for performing voice rule synthesis from character representations to which acoustic feature parameters are added. The voice output unit 110 is a unit that outputs voice, such as a speaker. The screen display unit 111 is a unit that performs control for displaying a document, parameters, and the like on the screen. The screen output unit 112 is a unit such as a display that outputs characters or graphics. The information synchronizing output means is means for synchronizing the respective output means when reproducing and outputting the voice information and the pointing information input at the same time. The sound reproducing means is means for reproducing and outputting the stored sound.

In the embodiment of FIG. 1, a document created by using the keyboard input means 100 or the mouse input means 101, the voice input means 104, and the voice recognition means 105 is analyzed by the document analysis means 102.

FIG. 2 shows, as the analysis contents of the document analysis means 102 in FIG. 1, (a) a document to be analyzed, and (b) a result of decomposing the document into constituent elements such as character strings and figures in the document. , (C) show examples of attribute values given to the respective components of (b). The document analysis unit 102 first decomposes the document into each constituent element (object) such as a character string, a figure, a numerical value, and a table in the document, and then, regarding each constituent element, a parent-child relationship, a class, a subclass, Analyze attribute values such as position coordinates in the document.

The attribute value as the analysis result of the document analysis means 102 is output to the presentation format conversion means 103 shown in FIG.

FIG. 3 is a diagram showing an embodiment of the presentation format conversion means 103 of FIG. In FIG. 3, a presentation format conversion means 301 is a means for converting a document into a sentence in a presentation format expression. As described above, the document analysis unit 300 adds the attribute value to the component group in the document and outputs it. Further, the presentation condition designating means 302 sets the conditions relating to the conversion method of the presentation format expression input using the keyboard input means 305 or the mouse input means 306 to the presentation format conversion means 30.
Output to 1.

[0012] Here, the condition relating to the method of converting the presentation format expression refers to a method of explaining a descriptive expression and a figure when explaining a title in a presentation. These are specified mainly by preparing selection candidates and allowing the user to select, but the user can freely change.

In the presentation format representation template 303, a presentation format representation template is stored in advance. Normally, the presentation format conversion unit 301 performs conversion processing on the document expression sent from the document analysis unit 300 by referring to the presentation format expression fixed form 303. However, if the user specifies the conversion processing method, the presentation format conversion means 301 performs the conversion processing based on the conversion conditions sent from the presentation condition specification means 302. In this way, the presentation format conversion unit 301 outputs the document representation converted into the presentation format representation to the parameter setting unit 304.

FIG. 4 shows the parameter setting means 107 of FIG.
It is a block diagram of an example of. In FIG. 4, a parameter setting unit 404 is a unit that adds a parameter for rule composition when reading a document in the presentation format expression to the output of the presentation format conversion unit 403. In FIG.
The presentation condition designating means 402 outputs to the parameter setting means 404 the conditions relating to the rule composition parameters input using the keyboard input means 400 or the mouse input means 401.

Here, the conditions relating to the parameter of rule synthesis are the time of presentation, the speed of utterance, the place and method to be emphasized in the presentation, how to add intonation, and the like. These mainly prepare selection candidates,
It is specified by the user's selection, but the user can freely change it. Further, among these conditions, the user can specify the utterance speed, the emphasis method, the intonation method, and the like, which can be obtained from the user's utterance, by the utterance.

Referring to FIG. 4, the voice recognition means 4
Reference numeral 06 performs voice recognition processing on the voice of the user input using the voice input means 405 such as a microphone. The phoneme duration calculation unit 407 calculates the phoneme duration based on the result of analysis by the voice recognition unit 406. The speech time measuring means 408 measures the speech time of the user. The power calculation means 409 calculates the power. The prosody information extraction means 410 extracts prosody information. The respective calculation results and extraction results are sent to the parameter setting means 404. The parameter setting means 404 adds the parameters thus input to the presentation format expression of the document which is the output of the above-mentioned presentation format conversion means 403, and outputs the result to the speech synthesis means 412.

FIG. 5 is a diagram showing the configuration of an embodiment of the parameter setting means 107 shown in FIG. In FIG. 5, the parameter setting means 107 is the parameter adding means 5
00, standard announcement time calculation means 501, announcement speed calculation means 502, announcement time calculation means 503, and standard phoneme duration time table 504.

The parameter adding means 500 adds the parameter extraction means 506 or the announcement condition designating means 50 to the announcement format representation of the document output from the announcement format conversion means 505.
The parameter output from 7 is added, and the parameter addition expression is output. Furthermore, the standard presentation time calculation means 50
1 calculates the time required for outputting the corresponding document as a presentation by speech for the parameter-added expression.

Here, when the standard presentation time calculation means 501 calculates the standard presentation time from the parameter-added expression, by referring to the standard phoneme duration table 504 in which the phoneme duration is stored in advance, The duration is checked for each phoneme in the additional expression, and the duration of all phonemes is added. Next, the standard presentation time calculation means 501 adds the pause lengths between all words, phrases, and sentences in the parameter addition expression added by the parameter addition means 500. Furthermore, the standard presentation time calculation means 501 adds the total phoneme duration time and the total pose length,
Calculate and output the standard publication time of a document.

Here, the flow of processing in the parameter setting means differs depending on whether the announcement time is designated in advance or when the announcement speed is designated.

FIG. 6 shows a flowchart of processing in the parameter setting means. Referring to FIG. 6 with reference to the block diagram of FIG. 5, when the standard presentation time is calculated by the standard presentation time calculation means 501 (s1), and the presentation time is specified by the presentation condition designating means 507 (s
2) The presentation speed calculation unit 502 expands or contracts the duration of each phoneme so that the standard presentation time and the specified presentation time match (s3). As a result, the determined phoneme duration is added to the parameter addition expression and output to the presentation time calculation unit 503. The presentation time calculation means 503 calculates the voice output time of each structural unit such as a page, paragraph, figure in the parameter addition expression (s4) and outputs it to the screen display means. Further, the parameter addition expression is output to the voice synthesizing means (s5).

If the presentation time is not designated by the presentation condition designating means 507 (s2), but the presentation speed is designated (s6), the presentation speed calculating means 502 is not used and the presentation time calculating means is not used. In 503, the voice output time for each constituent element such as page, paragraph, and figure in the parameter addition expression is calculated based on the phoneme duration length ratio obtained from the specified presentation speed (s7), and is displayed on the screen display means. Output. At this time, the designated presentation speed is added to the expression which is the output of the standard presentation time calculation means 501 and sent to the speech synthesis means (s5).

FIG. 7 shows the information synchronization output means 1 shown in FIG.
13 is a diagram showing the configuration of an example of No. 13 of FIG. Voice storage means 7
03 stores the user's voice input using the voice input means 700 such as a microphone. The pointing information storage means 704 is a pointing input means 70.
The pointing information input using 1 is stored.
The output synchronization unit 705 outputs the voice and pointing information that is input and recorded in advance corresponding to a part of the document, as necessary. Here, the voice storage means 7
03, the audio information is received and output to the audio reproducing means 706. Similarly, it receives the pointing information from the pointing information storage means 704 and outputs it to the screen display means 707. At that time, the output synchronization unit 705 controls the audio reproduction unit 706 and the screen display unit 707 so as to output these information in synchronization. If there is no pointing information input at the same time, the audio reproducing unit 706 outputs only the audio information.

In this way, the function of confirming the presentation is realized by analyzing the created document, converting the analyzed document into the presentation format expression, and reading the presentation format expression by voice synthesis. Further, at this time, a function of linking the voice and pointing information input by the user with the document and storing the document is realized.

Next, in FIG. 1, means used for the presentation evaluation function are voice input means 104, voice recognition means 105, parameter extraction means 106, parameter determination means 115, screen display means 111, and screen output means 112. .

The voice input means 104 is a tool such as a microphone used for inputting voice. The voice recognition unit 105 is a unit that analyzes the input voice and outputs a corresponding word string and other analysis results.
The parameter extraction unit 106 is a unit that extracts the acoustic feature parameter of the input voice. The parameter determining unit 115 is a unit that compares the value of the parameter extracted from the voice of the user with a standard value that is set in advance to make a determination. The screen display unit 111 is a unit that performs control for displaying a document, parameters, and the like on the screen. The screen output unit 112 is a unit such as a display that outputs characters or graphics.

In the embodiment of FIG. 1, voice input means 10
The user's voice input using 4 is the voice recognition means 1
It is recognized at 05, and the acoustic feature parameter of the voice of the user is extracted at the parameter extraction means 106.

FIG. 8 is a diagram showing the configuration of an embodiment of the parameter extracting means 106 shown in FIG. The parameter extraction unit 106 of FIG. 1 extracts various parameters of the user's voice in order to evaluate the user's presentation, and the parameter determination unit 115 makes a determination. In FIG. 8, the keyword is treated as a parameter extracted from the input voice. The keyword extracting unit 802 is a unit that extracts a keyword with respect to the result obtained by processing the voice of the user input using the voice input unit 800 by the voice recognition unit 801.

In FIG. 8, first, the voice of the user input using the voice input means 800 such as a microphone is sent to the voice recognition means 801, and acoustic analysis is performed. Keywords are extracted by the keyword extraction means 802 from the analyzed results, and the screen display means 803 controls the screen output means so that the document in which the keywords are inserted at a predetermined position is displayed on the screen.

FIG. 9 is a diagram showing the configuration of another embodiment of the parameter extracting means 106 shown in FIG. The keyword extracting unit 902 is a unit that extracts a keyword with respect to the result of processing by the voice recognition unit 901 the voice of the user input using the voice input unit 900. The keyword determining unit 904 is a unit that compares the input keywords with keywords prepared in advance and determines the keywords.

In FIG. 9, first, the voice of the user input using the voice input means 900 such as a microphone is sent to the voice recognition means 901 and acoustically analyzed. Keywords are extracted by the keyword extracting means 902 from the analyzed results. Here, the document analysis unit 903 decides to extract the keyword in the document, and the keyword determination unit 904 compares this keyword with the keyword as a result of the keyword extraction unit 904 to determine the matching ratio. And screen display means 905
The screen output means is controlled so that the result of this determination is displayed on the screen.

FIG. 10 shows the parameter extraction means 10 of FIG.
6 is a diagram showing the configuration of yet another embodiment of No. 6 of FIG. The parameter extraction unit 106 of FIG. 1 extracts various parameters of the user's voice in order to evaluate the user's presentation, and the parameter determination unit 115 makes a determination. Here, as the parameters to be determined, the phoneme detection rate, intonation,
It deals with speaking speed and speaking time. Phoneme detecting means 100
Reference numeral 1 is a means for detecting phonemes from the user's voice. The phoneme detection rate determination means 1002 is means for determining the proportion of the detected phonemes. The prosody information analysis means 1003 is means for analyzing prosody information from the voice of the user. The intonation determination means 1004 is means for determining the degree of intonation obtained from the analyzed prosody information. Speech rate analyzing means 100
Reference numeral 5 is a means for analyzing the speech rate from the voice of the user. The speech rate determining means 1006 is means for determining the analyzed speech rate. The utterance time measuring means 1007 is means for measuring the utterance time of the user. The utterance time determination means 1008 is means for determining the measured utterance time.

Here, as an example, the case of determining the speech rate will be described. First, the user's voice input using the voice input unit 1000 such as a microphone is sent to the voice recognition unit 1010 and acoustically analyzed. From the analysis result, the speech rate analyzing means 1005
Speech rate is analyzed at. The preset speech rate is compared with the speech rate as a result of the speech rate analysis unit 1005 to make a determination. And the screen display means 10
The screen output means is controlled by 09 to display the result of this determination on the screen.

In this way, the voice of the user is analyzed to determine features such as the phoneme detection rate, speech time, and intonation,
By evaluating, the function of presentation evaluation is realized.

[0035]

[Effects of the Invention] The created document is analyzed, and the analyzed document is converted into a presentation format expression and read aloud by voice synthesis, so that the user can confirm the presentation content and method at the same time as creating the original for presentation. Made it possible. In addition, by analyzing the voice of the user during the practice of the presentation and evaluating the acoustic characteristics such as the speech speed, the speech time, and the intonation, it is possible to evaluate the practice of the presentation of the user.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a presentation support device of the present invention.

FIG. 2 is a block diagram showing an embodiment of document analysis means.

FIG. 3 is a diagram showing an example of processing of presentation format conversion means.

FIG. 4 is a block diagram showing an embodiment of parameter setting means.

FIG. 5 is a block diagram showing the configuration of an embodiment of parameter setting means.

FIG. 6 is a flowchart of a process in a parameter setting unit.

FIG. 7 is a block diagram showing the configuration of an embodiment of an information synchronization output means.

FIG. 8 is a block diagram showing the configuration of an embodiment of parameter extracting means.

FIG. 9 is a block diagram showing the configuration of another embodiment of parameter extracting means.

FIG. 10 is a block diagram showing still another embodiment of the parameter extracting means.

[Explanation of Codes] 100 ... Keyboard input means, 101 ... Mouse input means, 102 ... Document analysis means, 103 ... Presentation format conversion means, 104 ... Voice input means, 105 ... Voice recognition means, 1
06 ... Parameter extracting means, 107 ... Parameter setting means, 108 ... Presentation condition designating means, 109 ... Voice synthesizing means, 110 ... Voice output means, 111 ... Screen display means, 1
12 ... Screen output means, 113 ... Information synchronization output means, 11
4 ... Voice reproducing means, 115 ... Parameter judging means.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G10L 3/00 551 Z 9/00 301 A (72) Inventor Yasumasa Matsuda Yoshida Totsuka-ku, Yokohama-shi, Kanagawa 292, Machi, Hitachi Ltd., Visual Media Research Laboratory (72) Inventor, Ojo Adult, 292, Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture, Japan, (72) Incorporated, Hitachi, Ltd. Visual Media Research Institute, Hasegawa, Yokohama, Kanagawa 292 Yoshida-cho, Totsuka-ku, Hitachi Ltd. Visual Media Research Laboratories

Claims (20)

[Claims] 1. At least a character input means such as a keyboard, a pointing input means such as a pen or a mouse, a document analysis means for analyzing a document including not only characters but also a chart, and a sentence in a sentence format of a presentation. The presentation format converting means for converting, the voice inputting means, the voice recognizing means for recognizing the voice of the user, the parameter extracting means for extracting the characteristic parameter of the voice of the user, and the parameter for converting the sentence into the voice Parameter setting means for setting, presentation condition converting means and announcement condition specifying means for designating the method of the parameter setting means, voice synthesizing means for converting sentences into voices, voice output means for outputting voices, documents, and Screen display means for displaying the characteristics relating to reading on the screen, screen output means, information synchronization output means, and audio reproduction means A confirmation device for assisting a presentation with at least said audio input means, the speech recognition means, and said parameter extraction unit, and the parameter determining means for determining characteristic parameters of the user's voice,
A presentation support device comprising a combination of the screen display means and a training device having the screen output means for assisting a presentation. 2. The confirmation device for assisting the presentation according to claim 1, which has a function of enabling the user to confirm the content of the presentation and the method of freely setting the conditions to be read out or by the conditions specified by the user. A presentation support device characterized in that 3. The training device for assisting a presentation according to claim 1, wherein the training support device has a function of evaluating the user's presentation by determining various characteristic parameters of the user's voice. . 4. The presentation support apparatus according to claim 2, wherein the document analysis unit has a function of analyzing attribute values such as type and position for each constituent element in the document. 5. The presentation support apparatus according to claim 2, wherein the presentation condition designating means converts the presentation format representation of the information entered by using character input means or pointing input means. A presentation support device having a function of outputting to a parameter setting means and a parameter of voice rule synthesis at the time of reading to the parameter setting means. 6. The presentation support apparatus according to claim 2, wherein the presentation format conversion means uses a presentation format expression fixed form prepared as a standard presentation format expression, and a document output by the document analysis means. A presentation support device having a function of converting a group of components into a presentation format expression. 7. The presentation support apparatus according to claim 2, wherein the presentation format conversion means presents a group of document constituent elements output by the document analysis means on the basis of a presentation condition output by the presentation condition designating means. A presentation support device having a function of converting into an expression. 8. The presentation support device according to claim 2, wherein the parameter extracting means calculates a phoneme duration length calculation means for calculating a phoneme duration time for a voice input by a user, particularly using a voice input means. , Utterance time measuring means for measuring utterance time, prosody information extracting means for extracting prosody information, and power calculating means for calculating power, and calculates and extracts phoneme duration, utterance time,
A presentation support device having a function of outputting information such as prosody information and power to a parameter setting means. 9. The presentation support apparatus according to claim 2, wherein the parameter setting unit acoustically outputs a presentation expression output by the presentation format conversion unit such as an emphasis method output by the presentation condition specifying unit and a pause length. A presentation support apparatus having a parameter addition means for adding characteristic information. 10. The presentation support apparatus according to claim 2, wherein the parameter adding means outputs an expression in which acoustic feature information is added to a character expression of the document, and the parameter setting means: A phoneme duration is added from the standard phoneme duration table in which the standard phoneme duration is stored, and all phoneme durations of the expression are added, and word-to-word or phrase-to-sentence-to-sentence A presentation support apparatus having a standard presentation time calculation means for calculating a standard publication time of a document by adding a pause length. 11. The presentation support apparatus according to claim 2, wherein the parameter setting means makes an announcement about the standard presentation time of the document output by the standard presentation time calculation means from the presentation time output by the presentation condition designating means. A presentation support device having a presentation speed calculation means for calculating a speed. 12. The presentation support apparatus according to claim 2, wherein the parameter setting means, based on the presentation speed output from the presentation speed calculation means, expresses the total output time of the expressions output from the parameter addition means. Alternatively, the presentation support apparatus is characterized in that it has a presentation time calculation means for calculating the presentation time for each structural unit and further outputting the presentation time to the screen display means. 13. The presentation support apparatus according to claim 2, wherein the presentation speed calculation means outputs an expression obtained by adding the duration of each phoneme to the expression output from the parameter adding means. Decided to do
A presentation support apparatus, wherein the voice synthesizing unit has a function of synthesizing a voice from the expression and outputting the voice to a voice output unit. 14. The presentation support apparatus according to claim 3, wherein the audio storage means stores audio information input using the audio input means, and the audio reproduction means reproduces and outputs the audio stored in the audio storage means. And
The pointing information storage means stores the pointing information input using the pointing input means, and the information synchronization output means stores the voice stored in the voice storage means and the information stored in the pointing information storage means. Regarding the above, the presentation support device is characterized by having an output synchronization means for instructing the audio reproduction means and the screen display means to output the information by synchronizing the information inputted at the same time. 15. The presentation support apparatus according to claim 3, wherein the parameter extraction unit analyzes the voice input using the voice input unit and extracts the keyword, and the keyword extraction unit extracts the keyword from the corresponding position of the document. And a screen display means for displaying the inserted document on the screen. 16. The presentation support apparatus according to claim 3, wherein the document analysis unit extracts and outputs a keyword in the document, and the parameter extraction unit collates the keyword and the keyword extraction unit. A presentation support device comprising: a keyword determination unit for determining a keyword output by the above, and the determination result is displayed on the screen by the screen display unit. 17. The presentation support device according to claim 3, wherein the parameter extraction means analyzes the voice input using the voice input means, performs a phoneme detection process, and outputs a phoneme detection rate. And a phoneme detection rate determination means for determining the phoneme detection rate, and a function of displaying the determination result on the screen in the screen display means. 18. The presentation support device according to claim 3, wherein the parameter extraction unit analyzes the voice input using the voice input unit, analyzes the prosody information, and outputs the degree of intonation. Having an analyzing means and an intonation judging means for judging by comparing the degree of intonation with a standard tongue degree which is held in advance, and having a function of displaying the judgment result on the screen by the screen displaying means. Presentation support device. 19. The presentation support apparatus according to claim 3, wherein the parameter extracting unit analyzes the voice input by using the voice input unit and analyzes the voice speed, and the voice speed analyzing unit and the voice speed. A presentation support device comprising: a speech rate determining means for determining, and the screen display means having a function of displaying the speech rate or the determination result on a screen. 20. The presentation support apparatus according to claim 3, wherein the parameter extracting means analyzes the voice input using the voice input means and measures the utterance time, and the utterance time measuring means. A presentation support device, comprising: a speech time determining means for determining whether the speech time or the determination result is displayed on the screen by the screen display means.