JPS62996A - Sentence editing apparatus for voice announcement - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of the Invention) The present invention relates to a text editing device for voice guidance used for, for example, automatic traffic information guidance, automatic guidance at stations and airports, automatic answering of telephone calls, etc. , relates to a text editing device for voice guidance used when there is a restriction on the number of characters in a sentence to be voice guided.

(Summary of the Invention) The present invention is such a text editing device for audio guidance that is capable of providing audio guidance with as much detail as possible and as much information as possible within a set number of characters.

(Prior art and its problems) In this type of conventional device, for example, in the case of traffic information, for one piece of information, "△△ route △
``From △ point'', ``To △△'', ``From △△ point'', ``
Stylized sentences based on fixed elements were used, such as "To point △△", [△△kml, "There is a traffic jam"], and sentences were edited by sequentially connecting a large number of these pieces of information. .

However, in the case of the conventional example with such a configuration, when the number of pieces of information increases and exceeds the set number of characters, the priority order set in advance for the information is followed, and items with lower priority are selected. I had no choice but to cut it unconditionally,
There was a drawback that the number of pieces of information was reduced. Therefore, it is possible to reduce the number of fixed elements that make up the information, but even when the number of information is small and the number of characters in the edited text is significantly less than the set number of characters, the content of the information can be simplified, Regardless of the amount of information, there was a drawback that only voice guidance was always available, which lacked specificity.

(Object of the invention) The present invention was made in view of the above circumstances, and it changes the sentence pattern to be edited according to the amount of information, and allows more information to be edited without exceeding the set number of characters. The purpose is to provide voice guidance in as much detail as possible.

(Structure and Effects of the Invention) In order to achieve such an object, the present invention provides fixed elements common to all information and ranks of importance, as shown in the correspondence diagram with the claims in FIG. an information storage means 1 for inputting and storing a large number of pieces of information constituted by variable elements unique to the information, with priorities set for each; and an expression sentence pattern for editing a sentence using the information from the information storage means 1. The sentence pattern storage means 2 stores a large number of the information with different numbers of variable elements, and the information from the information storage means 1 is incorporated into the expression sentence pattern from the sentence pattern storage means 2 to create sentences based on a large number of information. The editing means 3 to be edited compares the number of characters of the text edited by the editing means 3 with the set number of characters, and when the number of characters exceeds the set number of characters, the editing means 3 selects a text with lower importance among the variable elements of the information. a selection means 4 which compares the number of characters of a sentence edited by incorporating it into the reduced expression sentence pattern with a set number of characters, and selects and outputs a sentence when the number of characters of the sentence becomes less than the set number of characters by the comparison; and configure it.

According to this configuration, even if there is a lot of information, if the text is edited by incorporating that information into the beginning of the expression, and the number of characters in that sentence exceeds the set number of characters, it will contain more information than in the previously incorporated expression. The text is edited by incorporating less important variable elements of the text into a reduced expressive sentence pattern, and the number of characters in the edited text is compared with the set number of characters. As a result of the comparison, if the number of characters in the sentence is greater than the set number of characters, the text is further edited by incorporating less important variable elements into a reduced expression sentence pattern. In this way, information is incorporated into the expressive sentence pattern while gradually reducing variable elements with low importance, and the number of characters of the edited sentence is compared with the set number of characters, and when the number of characters is less than the set number of characters, the sentence is is selected as the text for audio guidance.

Therefore, when the amount of information is small, it is possible to edit the text in a format that incorporates a large amount of information for each piece of information by using an expressive sentence pattern with many variable elements, and conversely, when the amount of information is large, the edited text Even if the number of characters exceeds the set number of characters, by reducing the variable elements, the text can be edited by adopting an expression style that is appropriate for the amount of information without reducing the number of information, and overall the number of information can be reduced. Corresponding to each, users can now edit sentences by expressing them in as detailed a format as possible while fully using the set number of characters, and can now provide voice guidance with as much information as possible in as much detail as possible.

(Description of Examples) Hereinafter, the present invention will be described in detail based on examples shown in the drawings. FIG. 2 is a block diagram of a voice guidance text editing device according to an embodiment of the present invention, where ll is a provided information collection unit;
fla is an input operation unit such as a keyboard for inputting provided information to the provided information collection unit 11 using a code number or the like. 12 is the provided information collection department■
Each time new information is collected in step 1, a text editing section 12a edits the text in a predetermined expression pattern based on the provided information; This is an input operation unit such as a keyboard that performs.

Reference numeral 13 denotes an audio editing device, which reads audio data corresponding to the audio code from words, phrases, etc. stored in the audio source memory 14 based on the audio code from the text editing unit 12, and reads audio data corresponding to the audio code from the D/A converter 15...・It is designed to output to .

16 is an amplifying section, and 17 is an audio providing section such as a speaker, telephone, or wireless transmitter.

The provided information collection section 11, text editing section 12, and audio editing device 13 are each composed of a computer. It should be noted that various methods may be used, such as configuring the entire system on one computer, or configuring the provided information collection section 11 and text editing section 12 on one computer, and the audio editing device 13 on another computer. Configurable.

In the text editing section 12, the information storage means 1;
A sentence pattern storage means 2, an editing means 3, and a selection means 4 are provided.

In the information storage means 1, as shown in FIG. 3, n pieces of information are input and stored in an information table T+.

These information l, 2. ..., (n-1), n are an element P unique to all information and a variable element Ai unique to information, respectively.

Bi, Ci, ---, Ni, Oi (however, f: l,
2, =-, n-1, n).

Among these variable elements, Ai and Bi are the most important variable elements and are the minimum necessary elements that cannot be omitted. The other variable elements Ci, . . . , Ni, Oi are ranked in order of importance so that C1>Di>..>Ni>Of. In addition, these information l, 2. ..., (n-1), n is information 1>information 2
The priority order is set in the order of >->information (n-1)>information n.

In the sentence pattern storage means 2, as shown in FIG. 4, m expressive sentence patterns are stored in a sentence pattern table T for text editing.

These expression sentence types 1.2. ..., (m −1), +1
1 is the variable element Ai, Bi, Ci, . . . of the information.
From the most extended expression sentence l that contains n pieces of information consisting of all of Ni and Of and elements P unique to all information, n pieces of the most important variable elements At and Bi and elements P unique to all information. The variable elements are reduced one by one in descending order of importance until reaching the shortest expressive sentence pattern m constructed from m, and the sentence length is set to become shorter.

In addition, in the most shortened expression sentence type m, the most important variable elements Ai and Bi are repeated n times, and finally an element P unique to all information is added.

The editing means 3 is an information storage means! Information from 1.2
．． ..., (n-1), n is the expression sentence pattern l from the sentence pattern storage means 2, 2. . . , (m-1), m is incorporated into one expression sentence pattern to edit a sentence based on a large number of pieces of information.

The selection means 4 compares the number of characters of the text edited by the editing means 3 with the set number of characters, and when the set number of characters is exceeded, the editing means 3 reduces less important variable elements of the information. The number of characters in the text edited by incorporating it into the expression pattern is compared with the set number of characters, and based on these comparisons, the text is selected and output when the number of characters in the sentence is less than the set number of characters, and then , its detailed operation will be explained based on the flowchart of FIG.

First, given n pieces of information I, 2. ..., (n-1
).

The sentence is edited by incorporating n into the shortest expression sentence pattern l (Pi). It is compared and determined whether the number of characters Tm of this edited sentence is less than a preset number of characters (for example, about 300 characters) TIIlaX (F2). Here, if the number of characters is greater than the set number of characters T max, the process moves to step F3, cuts the information with the lowest priority among the information, returns to step PI, and compares and judges the number of characters with the set number of characters T max. . In this way, step F! Then, in step F3, first, the number of information that falls within the set number of characters T max is determined in advance.

In step F2, when it is determined that the edited number of characters Tm has become less than the set number of characters Tmax, the process moves to step F4, and the most extended type of expression pattern 1 is assigned as the expression pattern e in which information should be incorporated. Next, the sentence is edited by incorporating the number of information determined by the above processing into the expression sentence type I (F5). Thereafter, it is compared and determined whether the number of characters TQ of the edited text is less than the set number of characters T max (F6).

Here, when the number of characters is greater than the set number of characters T max, the most extended expression sentence pattern 1 is set as the expression sentence pattern ρ in which information should be incorporated.
The next shortest expression type 2 is assigned (F7), the sentence is edited using the expression type 2, and the number of characters is compared and determined in step F6. By repeating this, when the number of characters becomes less than the set number of characters T max, the expression sentence type at that time is adopted as the expression sentence type e in which information should be incorporated, and is output from the text editing section 12 to the audio editing device 13 (F
8).

After this, all the information-specific elements P and the information-specific variable elements At, Bi, in each of the information incorporated into the expression sentence type g from the words, phrases, etc. stored in the sound source memory 14,
...Audio data corresponding to each audio code is read and transmitted from the D/A conversion section 15... to the amplification section 16...
Voice guidance is provided through a voice providing section 17 such as a speaker, telephone, or wireless transmitter.

Next, the above-mentioned text editing will be specifically explained using the case of traffic information as an example.

As the most important variable element At, the sentence/clause "from △△ point of △△ route" is set, and as the other most important variable element Bi, the word [△△kmJ] is set. Furthermore, as the variable element Ci, “bound to △△”
However, the variable element Di is ``through point △△'', and the variable element Ei is ``to point △△'', and each variable element Fi, ..., Ol is a predetermined phrase or word. is set. The phrase "There is a traffic jam" is set as the element P unique to all information.

This allows, for example, traffic management centers, etc.
While looking at the various information that comes in, such as traffic jam information, weather information, accident/construction information, etc., the operator inputs the information as appropriate to the provided information collection section 11, and then transfers that information to the text editing section 12. Input and store in table T. Thereafter, the aforementioned selection means 4 performs processing to edit the text according to a predetermined expression pattern.

For example, if the shortest expression sentence m is adopted and all the information is incorporated, then "Δ△km from Δ△ point of △△ route, Δ△km from △△ point of △△ route"
△km...,...△△k111 There is a traffic jam.

' is edited.

In addition, if the expression sentence pattern (m-3) is adopted and all the information is incorporated, then ``From △△ point of △△ route to △△ point △△km Δ△bound through △△ point to △△ point. There is a traffic jam, △△km from the △△ point on the △△ route.
There is a traffic jam going from △△ to △△ point via ΔΔ point.
・・・、・・・There is a traffic jam up to △△ point. ' is edited.

Needless to say, if the amount of information decreases, the text can be edited by increasing the number of variable elements within the set number of characters.

In the above embodiment, in steps F1 to F3, first, the number of information that fits within the set number of characters T max is determined in advance, and a judgment is made as to whether the number of information is appropriate or not, with the number of edited characters being small, As a result, for example, if information is incorporated into the most extended expression sentence pattern l from the beginning and then sequentially shortened expression sentence patterns are assigned, a comparison judgment is made with the set number of characters up to the shortest expression sentence pattern m. rear,
Cut the lowest information and use the remaining information as an expression sentence pattern
In extreme cases, it is necessary to repeatedly incorporate information from expression pattern l to the shortest expression pattern m and repeat comparison judgments over and over again. , has the advantage of being able to satisfactorily shorten the processing time required for expression pattern selection, but in the present invention, step F! The process of step F3 may be omitted, and a process flow for cutting the number of pieces of information may be provided after step F7.

[Brief explanation of drawings]

FIG. 1 is a diagram showing how the text editing device for voice guidance of the present invention corresponds to the claims, FIG. 2 is a block diagram showing an embodiment of the text editing device for voice guidance of the present invention, and FIG. FIG. 4 is a diagram showing an information table, FIG. 4 is a diagram showing a sentence pattern table, and FIG. 5 is a flowchart. l...information storage means, 2...sentence pattern storage means, 3...editing means, 4...selection means.

Claims (1)

[Claims] (1) Information storage means for inputting and storing a large number of pieces of information consisting of fixed elements common to all information and variable elements unique to the information ranked in importance, with priorities set for each piece of information; Sentence pattern storage means for storing a large number of expression sentence patterns for editing sentences using information from the information storage means, with different numbers of variable elements of the information; an editing means that edits a sentence using a large amount of information by incorporating it into a sentence pattern; and comparing the number of characters of the sentence edited by the editing means with a set number of characters, and when the number of characters exceeds the set number of characters, the editing means edits a variable element of the information. Compare the number of characters of a sentence edited by incorporating the less important ones into an expressive sentence pattern with the set number of characters, and by comparing them, calculate the sentence when the number of characters is less than the set number of characters. A text editing device for audio guidance, comprising a selection means for selectively outputting.