JP3712325B2 - Document reading device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a document reading apparatus, and more specifically, when reading a document along the document structure, the structure of each structure can be understood so that the structure of the inclusion relationship or parallel relationship in the text document is read out by speech synthesis. A voice or sound for guide is added to the part or its delimiter so that the listener can understand the contents following the change of the topic in the document.
[0002]
[Prior art]
Usually, a document is separated into a plurality of parts by applying a visual layout. And the structure of inclusion relation or parallel relation is made by the positional relation of each part, and the strength of the relation is expressed. On the other hand, with current text-to-speech synthesis, you can simply read a document monotonously and read it at best by changing the sound quality by character modification, or improve the sound quality for the quoted range explicitly indicated by signals etc. in e-mail etc. However, it is not possible to express related information between documents expressed in a layout.
[0003]
[Problems to be solved by the invention]
The present invention has been made in view of the above circumstances, and when a document structured by layout information is read out by text-to-speech synthesis, the target document is hierarchized according to its inclusion relationship etc. The purpose is to help recognition and support understanding of document contents, and to make it easier for the listener to understand the contents by using structural information not displayed on the screen.
[0004]
[Means for Solving the Problems]
The invention according to claim 1 is a document reading device that reads out a structured document by regular speech synthesis, and includes a guide speech and / or acoustic data storage device, a speech synthesis device and / or an acoustic signal generation device, This is characterized in that the delimitation of the document structure of the document is expressed by a guide voice.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a principal block diagram for explaining an example of a document reading apparatus to which the present invention is applied. In the figure, 1 is a data input device to which structured document data is input, and 2 is from the data input device 1. Data structure analysis device for analyzing the structure of input data, 3 is a data storage device, 4 is a voice output control device, 5 is a guide voice / acoustic data storage device, 6 is an input device operated by a user, and 7 is a voice A synthesizing device, 8 is an acoustic signal generating device, 9 is an output device that utters sounds such as speakers and buzzers, and the present invention reads out a structured document input from the data input device 1 along the document structure, When outputting an audio from the output device, a structure sound is clarified by inserting a guide sound or sound into the structure break according to a request from the input device.
[0015]
As described above, according to the present invention, when reading a document along the document structure, the section of the structured document is clarified so that the listener can recognize the document structure and understand the contents of the document well. However, examples of a method for conveying the document structure by sound or voice include a method for clarifying the division of the document structure and a method for expressing the relationship between items.
[0016]
To clarify document structure delimiters:
・ Sound is added to the structure breaks.
Add sounds such as “pea” or “boo” once or multiple times in a continuous or single shot.
・ Insert a silent part.
・ Enter a guidance voice according to the distance between each topic item.
-Overlay the acoustic signal on the background of the entire item.
The background sound is superimposed on an item in a certain hierarchy, and when the item changes, the background sound is also changed to know the change of the reading item.
[0017]
How to express relationships between items:
・ This is a change in the sound of the break.
Change the length of the sound depending on the degree of relevance. For example, the thing with a weak relationship lengthens time.
Change the number of sounds. Insert a short sound several times, and increase the number of times when the relationship is weak.
・ Change the sound interval. For example, it is assumed that a certain number of sounds are input, and the interval is increased when the relationship is reduced.
・ Change the pitch of the sound. If the relationship between the front and rear is low, the sound is also low.
-Insert silence. The silent time is changed according to the degree of the relationship between before and after.
・ A combination of the above.
[0018]
・ Express with guide voice.
Prepare a different sound quality than the one used to read the text, and guide you when the topic changes significantly. For example, phrases such as “the story changes” and “the next topic” are determined for each hierarchy, and the corresponding phrase is inserted when an item in the hierarchy changes.
In bulleted items, if the item number is not numbered or there are multiple paragraphs, it will guide you to what number.
[0019]
-Know the depth of the general hierarchy by BGM and sound.
For example, the pitch and speed are changed according to the depth of the hierarchy. For example, when the hierarchy becomes deep, the speed of BGM or sound is changed. Since the BGM is the same, the range of reading is not changed, but it can be used when moving to detailed items.
Change the pitch of the acoustic signal.
[0020]
(Example 1)
FIG. 2 is a diagram showing an example of a document structure to which the present invention is applied (an example of a newspaper article). Hereinafter, an example in which the pitch of a sound signal and the direction of change are expressed in a narrow or narrow manner will be described. The inside is the sound and the <> is the meaning of the signal.
Politics (pi) <Move to lower level>
Election (pi) <Move to lower level>
Liberal Democratic Party Activities (P) Democratic Party Activities (P) <Parallel items are lined up>
Social Democratic Party Activities (Pu) Communist Party Activities (Buo) <Move to Higher Level>
Government Movement (Boo) <Up to higher levels>
Economy (pi) <lower level>
Trends in yen depreciation (P) <Items on the same level>
Stock Price Trends (Boo) <To higher rank>
Sports [0021]
(Example 2)
FIG. 3 is a diagram showing another example of the document structure to which the present invention is applied (example of data catalog). Hereinafter, an example in which the number of sounds and the direction of change or change direction will be described.
Food (pipi) <Move to lower level>
Soft drink (pipi) <lower level>
Carbonated drink (pi) <lined items line up>
Fruit juice (pipipipii) <up to higher level>
Confectionery (upper layer)
Daily necessities (pipi) <Move to lower level>
Kitchenware (pi)
Laundry goods (pi)
Bath products (upper layer)
Clothing (pipi) <Move to lower level>
Children's clothes (pi)
Women's clothing (pi)
Men's clothes (pipipipi)
[0022]
(Example 3)
A description will be given of an example in which a hierarchy is moved by the type and pitch of background music (BGM) using the example of the catalog shown in FIG.
Grocery (pipi) (same music plays in the background during the same field)
Soft drink (the pitch of background music increases) <to lower level>
Carbonated drink (pi) <lined items line up>
Fruit juice beverage (background music pitch is lowered)
Sweets (Back to the beginning of the background music pitch)
Daily necessities (pipi) (background music changes to another)
Kitchen utensils (background music pitch increases) <Move to lower level>
Laundry goods (pi)
Bath products (background music pitch is lowered) <upper level>
Clothing (pipi) (background music changes to another)
[0023]
(Example 4)
FIG. 4 is a diagram showing another example of a document structure to which the present invention is applied (an example of a magazine article). Hereinafter, an example using guide voice and sound will be described. , Acoustic.
(First topic is) Politics (First item is) Pledge of each party LDP's pledge (Phi Phi), Democratic pledge (Phi Phi), Communist Party pledge (Boo)
(The second item) Predicting seats to be won (the story has changed)
Sports information (first item) World Cup (second item) Professional baseball XX consecutive losses (third item) Sumo (the story has changed)
Entertainment information [0024]
【The invention's effect】
As is clear from the above description, according to the present invention, when a structured document is read out along the document structure, the division of the document structure is clarified by sound or sound, or the relationship between items is sounded or guided. Can be clarified by voice.
[Brief description of the drawings]
FIG. 1 is a principal block diagram for explaining an example of a document reading apparatus to which the present invention is applied;
FIG. 2 is a diagram showing an example of a document structure to which the present invention is applied (an example of a newspaper article).
FIG. 3 is a diagram showing another example of document structure to which the present invention is applied (example of data catalog).
FIG. 4 is a diagram showing another example of a document structure to which the present invention is applied (an example of a magazine article).
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Data input device, 2 ... Data structure analyzer, 3 ... Data storage device, 4 ... Voice output control device, 5 ... Guide voice / acoustic data storage device, 6 ... Input device, 7 ... Speech synthesizer, 8 ... Sound Signal generator, 9... Output device.

Claims (1)

A document reading device for reading a structured document by regular speech synthesis, comprising a guide voice and / or acoustic data storage device, a speech synthesis device and / or an acoustic signal generator, and delimiting the document structure of the structured document A text-to-speech device characterized in that it is expressed by a guide voice.

Images