JPS5922080A - Voice news - 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 an audio newspaper that can be easily read even by visually impaired people.

Structure of the conventional example and its problems Conventional books for the blind include Braille books using perforations and books using foam ink. Blind people read Braille information by touch, but the amount of information that can be stored per page is extremely small, and the reading speed cannot be said to be fast. Such Braille books are difficult to make copies, relying on volunteer work and requiring special copying equipment.

It is significant for visually impaired people to encode sounds and print them as patterns in large quantities using normal printing means and sell them. It is extremely inconvenient to be able to detect the pattern of sounds only after several haphazard trials in which it is impossible to know exactly where (r)l'l is.

Purpose of the Invention The present invention eliminates the above-mentioned drawbacks, and by distributing audio information in print, provides blind people with a large amount of information inexpensively, quickly, and at any time and place that is available in printed matter. The aim is to provide audio newspapers and audio books that take advantage of their readability characteristics and allow even visually impaired people to easily locate audio patterns.

Structure of the Invention The present invention encodes audio information, converts the code into a code, prints it out using an ordinary printing machine, copying machine, facsimile machine, etc., reads the printed code with a reading device, and converts the code into a code. This code is added to the speech synthesizer, and when providing the audio information as printed matter, the system that reproduces the audio can identify the location and state of the audio information with the printed audio information. Related things include useless things that make up newspapers, books, etc.

The printing pattern according to the present invention is applicable not only to conventionally known barcodes, but also to mosaic-like patterns in which codes are expressed by changing at approximately equal density in each direction of two orthogonal coordinate axes. can. In this case, 1 ct
Information description capacity per M is 1600 bytes/c'A to 1
Reach 0000 pits/CA.

DESCRIPTION OF EMBODIMENTS The present invention will be described below based on embodiments in which the print pattern is mosaic-like. However, in addition to this example, the present invention can also be applied to print patterns such as barcodes.

Figure 1a shows an audio newspaper for the blind with an audio printing pattern except for special identification information such as headings and dates.
It is. In FIG. 1, a rectangle 11 represents one block of the audio pattern. Between the blocks there is a blank area 13 with no audio pattern. 1
The width of 3 is appropriate. As is clear from FIG. 1, the block 11 is surrounded by a blank area 14, that is, a non-print pattern area 14. Although the above description has been made using a white background, the situation is the same even if the white background is easily distinguishable from a printed pattern, such as a black background or a fine line drawing.

The voice pattern is read using a reading device which will be explained later with reference to FIG. 6, but a visually impaired person cannot know the position of the block of the voice pattern.

What is shown in FIG. 1 815 is a marker or a hole provided at the beginning of each block on the newspaper surface, corresponding to each block. Such a device remains constant even if the content of audio information changes as long as the shape and position of one block are fixed, and the device for this purpose may be simple. By touching this marker 16, the position of the audio block can be known. Although FIG. 1 shows examples of various types of markers, the present invention is not limited thereto, and includes any marker associated with an audio block. In FIG. 1, 10B is an audio newspaper and 11B is an audio pattern block. 15B is one marker hole, and this is the case where there are a plurality of blocks in one direction, and each block has a plurality of holes 15 as markers. 16A is a defective marker placed on the edge of a newspaper, 16B is a marker that can be felt with the finger such as foam ink or thick ink, and objects with magnetic materials, dielectric materials, visible light, or special colors other than visible light. , a marker made of metal, etc. A sensor is installed in the reading device except when using finger touch. 17 is a reading device, which scans the center of the block and scans the upper part 1.
In order to ensure that the reading device 7A and the lower portion 17B are approximately equal to each other and are removed from the block, a tone signal or the like is used to notify when the reading device is removed from the block.

FIG. 2 is an enlarged view 20 of the portion surrounded by the small circle 12 in FIG. 1. In FIG. 2, considering two coordinate axes that are substantially orthogonal on a plane, for convenience, an arrow 24 indicates the X direction, and an arrow 26 indicates the Y direction, and the direction in which each arrow points is assumed to be positive. 21A and 21B are audio/
A guard band is used to determine the block width determined by the number of stitches in the Y direction of the cittern block and the positions of both ends of the block.It is a guard band that is made by arranging black stitches in the HxX direction. , and other color combinations. Here, an eye is a small area on the image corresponding to one bit of the code, and the position relative to the other eyes is The location is determined,
In my eyes, if the code to be converted into each turn is binary, the color does not change. The shape of the eyes is square in Figure 2, but
It is only necessary to know whether the pattern at that position means 0 or 1, and it may be of any other shape such as a circle or a triangle. Color refers to the color of the eyes, and binary signals 0 and 1 are used in correspondence with codes, such as 0 for white and 1 for black.

Any combination of color and code may be used.

The printed pattern may be made of ink other than black, as long as it produces contrast on the paper surface, such as when using full-length photographic film or colored paper used for facsimile. The contrast referred to here is not necessarily limited to visible light, but may also be contrast to ultraviolet, infrared, electromagnetic waves, and ultrasonic waves. Although the guard band for the block width in the Y direction has been described, the same applies to the X direction. 22A and 22B are phase bands. Phase bands are used to make it easier to know the position of each stitch in a pattern, and within one professional band, for example, inside the guard band, they are arranged in the X direction, and each stitch is colored in the X direction. This is to reverse the . When reading by moving the reader in the X direction, the phase band can be provided on both sides or one side of the block or at any other suitable location along the X direction. Furthermore, for convenience, the arrangement of eyes in the X direction will be called a row, and the arrangement of eyes in the Y direction will be called a column. The pattern portion sandwiched between rows 22A and 22B is a pattern portion corresponding to a code consisting of an arbitrary combination of a voice code, an error detection code, and an error correction code.

The following three types of audio codes are used: The first method is called waveform recording. The audio waveform is quantized and encoded, and even if DPOM or other code compression means are used in part, the audio quality is good because it is a so-called waveform recording. However, the number of codes required per second is as large as 10 to sok bits/second. The second method is called analysis and synthesis, in which the characteristics of the sound are replaced with parameters and encoded. The typical method is PAR
There is COR. Characteristics of the speaker, timbre intonation, pauses, tone, etc. can be encoded, and the amount of code for one second of p is about 2 bits/second, and the original sound can be reproduced with fairly good quality. The third method is called rule synthesis, in which sounds are coded and encoded. Since the chords and sounds correspond to each other, the amount of code per second is about 50 bits/second, but it is only possible to speak in a certain way with a certain voice, which is extremely lacking in naturalness.

According to the present invention, since the recording density per unit area is high,
Not only can it be used to reproduce sounds using the rule synthesis method, but it also enables the reproduction of natural-sounding sounds over a considerable period of time using the analytical synthesis method.Although the playback time is shorter, waveform recording is also within the usable range. It is in. Error detection codes include parity codes and CRC codes, and these codes are used to check whether there are any errors in the reproduced code. If an error is found, that part is discarded or the part is processed according to predetermined rules. Therefore, another code is used instead. Error correction codes include ECC codes and the like, and if an error occurs in the code, it is automatically corrected within a certain range. Reference numeral 23 in FIG. 2 is a blank area between blocks, which corresponds to 13 in FIG. 1.

FIG. 3 is an explanatory diagram of a case other than a newspaper, in which blocks of audio patterns are mixed with characters and graphics. 3o is, for example, one page of a book. In the page 30, a division/graphic section 31 and a voice pattern section 32 are mixed. 33 is a voice pattern block, as shown in Figure 1.
This is explained using FIG. 2. As shown in the diagram, the voice pattern block and the text etc. portion can be mixed in such a way that the block portion and the text etc. portion can be largely separated, or they can be mixed such that the voice block is inserted between the lines of the characters. It's okay. Even in such a case, the marker 34 is provided as explained using FIG.

Next, as an example, in the case of analysis and synthesis, the process from utterance to printed matter and the process from reading the print pattern with a reading device to reproducing the utterance will be described. FIG. 4 is a diagram illustrating the process from voice to the creation of a printed voice pattern on the page of a book, newspaper, etc., and the voice signal from the microphone 4o is transmitted to the voice parameter analyzer 4.
1, and is converted into a code string based on audio parameters by an encoding device 42. As already mentioned, various codes required in the reproduction process are added to this code. This code is rearranged by the patterning device 43 into the same arrangement as the butter, and once it is stored in the memory device 44.
is accumulated in Reference numeral 46 denotes a memory device that stores document information such as characters. These codes are combined by the editing device 46. Thereafter, the printing manuscript creation device 47 creates a manuscript such as a voice pattern and characters on a printing plate. By printing this using the printing machine 48, a printed audio book 49 such as a book is created. FIG. 5 is a diagram illustrating the process of reading a printed sound pattern and restoring the sound, in which a part 60 of the sound pattern is imaged as an image 63 on the photoelectric conversion element 52 such as an image sensor 610CCD array. ,
It is read by a photoelectric conversion element. This signal is restored by a pattern decoding device 64 into a code corresponding to the code created by the patterning device 43 in FIG. This is temporarily stored in the audio parameter storage device 56, the code is read out from this at a predetermined speed, the audio synthesizer 66 converts it into an audio signal, and the speaker 67 is sounded to reproduce the audio.

Effects of the Invention According to the present invention, a means for encoding voice information, easily making large numbers of copies using printed matter as a medium, and transmitting information is provided. When using a digital reading device, it is generally preferred to indicate the position of the audio information by means of a marker, making it possible to trace the audio block. Furthermore, even when the reading device is mechanical and automatically traces, the present invention is not necessarily limited to blind people, in that markers such as those described above make it easier to read audio blocks.

[Brief Description of the Drawings] Figure 1a is a diagram explaining the audio newspaper according to the present invention, Figure 1 is a diagram explaining various markers, and Figure 2 is an enlarged explanation of the audio pattern in Figure 1. Figure 3 is a diagram explaining an example of applying the present invention to a book, Figure 4 is a diagram explaining the process of creating a voice pattern, and Figure 5 is a diagram explaining the process of reproducing voice from the voice pattern. It is a diagram. 10... Audio newspaper, 11... One block of audio pattern, 12... Small circle indicating area, 13... White background between blocks, 14...・・・・・・
・Surrounding white background, 15... hole marker, 10B
... Audio newspaper, 11B... One block of audio pattern, 15B... Marker with multiple holes, 16A, 1eB. 16C... Various markers, 17... Reader, 17A, 17B... Both ends of the reader,
20... Enlarged view of one voice pattern, 21A, 21B...
...Card pan)", 22A, 22B...fu, -zuno (nd, 23...white background between blocks, 24
...Arrow indicating the X direction, 26...Arrow indicating the Y direction, 30...One page of a book, 31...
...Character/graphic part, 32...Speech pattern part,
33...audio) ζ turn block, 34...
・・Marker, 40・・Microphone, 41・
...Audio parameter analysis device, 42...
Encoding device, 43...Patterning device, 44 8
Memory device for eyebrow pattern code, 46... Memory device for characters, etc., 46... Editing device, 47...
... Printing manuscript creation device, 48 ... Printing machine, 49
...Print audio book, 60 ...Part of sound pattern, 61 ...Image sensor-152...
Photoelectric conversion element, 63...Audio imaged on the photoelectric conversion element (turn, 64...Pattern decoding device, 56...Audio) (Rameter storage device, 66...
...Speech synthesizer, 57...Speaker. Figure 1 (α/4 Figure 1 (b) 5B Figure 3 Figure 4 Figure 5

Claims (1)

[Claims] An audio newspaper characterized in that audio information is encoded, the code is printed as a pattern, and means is provided for indicating the position of the pattern.