JP2000099624A - Device for reading text including image character - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the easiness of the understanding of character expression with an image by outputting a character expressed with an image as a synthetic voice. SOLUTION: Character codes and image codes in a text are separated by a character code/image separating means 1, and the separated image characters are recognized, and the recognized characters are replaced with the character codes by a character recognizing and character encoding means 2. Then, the character coded image characters and the original character codes are arranged according to the character array of the text by a reading-out order controlling part 3, and the order-controlled character codes are voice-outputted by a voice synthesizing means 4. Thus, even when the image characters are included in the text, the image characters can be character-encoded and read out.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a text-to-speech apparatus for reading text including image characters, and more particularly to a text-to-speech device including characters represented by images (image characters represented by bitmaps and vector data). The present invention relates to a text-to-speech apparatus for reading text including image characters which can be read aloud by synthetic speech.

[0002]

2. Description of the Related Art As electronic documents such as electronic mails and electronic bulletin boards have become more widely used for communication,
In some cases, the contents of digitized documents are confirmed by synthesized speech, such as access from outside by telephone or use by the visually impaired. Thus, on a homepage or the like, characters are often represented by images instead of character codes in order to enhance expressive power, but this cannot be converted into speech as it is.

[0003]

As described above, characters are represented by images on homepages and the like, but cannot be converted into voices using image characters as they are.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to make it possible to output a character represented by an image as synthesized speech, thereby improving the understandability of the character expression by the image. It was done as.

[0005]

According to the first aspect of the present invention, there is provided a character code / image separating means for separating a character code and an image character in a text, a character code / image separating means for recognizing the separated image character, and converting the recognized character to a character. Character recognition / character encoding means for replacing with a code, reading order control means for arranging the character-encoded image characters and the original character code in accordance with the character arrangement of the text, and speech synthesis of the order-controlled character code Voice synthesizing means for converting image characters into character codes and reading them out.

According to a second aspect of the present invention, in the first aspect of the present invention, there is provided a visual importance determining means for determining a visual attention degree of the separated image characters, and The reading order is controlled accordingly.

In a third aspect of the present invention, in the first aspect of the present invention, there is provided a visual importance determining means for determining a visual attention degree of the separated image characters, and Accordingly, the speech synthesis unit is controlled to change the auditory attribute of the separated image character portion.

According to a fourth aspect of the present invention, in the second or third aspect, a size of an image character is measured as the visual attention degree.

The invention of claim 5 is characterized in that, in the invention of claim 2 or 3, the contrast between the color of the image character and the background color is measured as the degree of visual attention.

A sixth aspect of the present invention is characterized in that, in the second or third aspect, a shape characteristic of a character is measured as the degree of visual attention.

According to a seventh aspect of the present invention, in any one of the first to sixth aspects, even if a single image is set, coordinate information for dividing the image is set to a plurality of corresponding areas. It is characterized in that it is handled separately as independent character information.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention is to output characters represented by images on a homepage or the like as synthesized speech, and is specifically implemented in the following form. 1) A text-to-speech conversion system that reads out text by recognizing characters by replacing them with character codes. 1.1) The degree of visual attention of characters is also determined and reflected in the reading order and auditory attributes. 1.1.1) Character size is used as the degree of attention. 1.1.2) The relationship between the character color and the background color is used as the degree of attention. 1.1.3) Character shape characteristics are used as the degree of attention. 1.2) Speech is output with auditory attributes different from those of the original character code text. 1.3) If coordinate information for dividing a single image is set, it is treated as independent character information for a plurality of corresponding areas.

FIG. 1 is a block diagram of a main part for explaining an example of a text-to-speech apparatus according to the present invention. In the drawing, reference numeral 1 denotes a character code (a character representing a character) in a text.
A code / image separation unit 2 for separating characters from an image (a part where characters are represented by images); an image character recognition / character coding unit 3 for recognizing characters of the separated image and coding the recognized characters; A reading order control unit for arranging the character codes separated by the code / image separation unit 1 and the character codes coded by the image character recognition / character coding unit 2 according to the text,
Reference numeral 4 denotes a speech synthesis unit that synthesizes the character codes from the reading order control unit 3 and outputs the synthesized speech.

As described above, according to the present invention, a character code and an image character are separated from a text in which a character code and an image character are mixed, and the image character is recognized and converted into a character code. It is designed to be output as a synthesized voice in combination with the character code that existed in the above. Furthermore, the visual attention level of the image characters separated as described above is also determined and reflected in the reading order and auditory attributes Let it. For example, the character size measuring unit 5 measures the size of the image character, the character color contrast measuring unit 6 measures the contrast between the character color and the background color, and further, the character shape similarity measuring unit 7 , The character characteristics such as the line thickness and dissimilarity with the standard font are measured, and the importance determining unit 8 determines the most important visual attention among them, and the reading order control unit 3 Controlling or changing the sound output by the sound synthesis unit 4 from the hearing attribute setting unit 9 to output a sound with a hearing attribute different from the text part by the original character code.

FIG. 2 is a diagram showing an example in which, even when a single image is set, coordinate information for dividing the image is handled as independent character information for a plurality of corresponding areas. is there. In other words, in the processing of the image map, the image is divided according to the accompanying coordinate information, thereby reading out as an independent menu item,
Independent link destinations can be set.

For example, when reading out an image map consisting of one image file and several lines of script as shown in Table 1, three regions indicated by three lines of <AREA...> (Dotted lines in FIG. 2) ) Are processed as independent images, and separate character strings of “previous page”, “next page”, and “return to table of contents” are extracted.

[0017]

[Table 1]

[0018]

As described above, characters are represented by images on homepages and the like, but cannot be converted into voices using image characters as they are. The present invention makes it possible to output a character represented by an image as described above as a synthesized voice, thereby improving the understandability of the character expression by the image.

[Brief description of the drawings]

FIG. 1 is a main part configuration diagram for explaining an embodiment of a text-to-speech apparatus including image characters according to the present invention.

FIG. 2 is a diagram illustrating an example in which a single image is divided into independent character information of a plurality of corresponding areas.

[Explanation of symbols]

1 code / image separation unit 2 image character recognition
Character encoding section, 3 ... reading order control section, 4 ... voice synthesis section, 5 ... character size measuring section, 6 ... character color contrast measuring section, 7 ... character shape similarity measuring section, 8 ... importance determining section, 9 ... Auditory attribute setting unit.

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Hiroo Kitagawa 1-3-6 Nakamagome, Ota-ku, Tokyo Stock inside Ricoh Company (72) Inventor Takashi Ariyoshi 1-3-6 Nakamagome, Ota-ku, Tokyo Stock Inside Ricoh Company (72) Inventor Yuichi Kojima 1-3-6 Nakamagome, Ota-ku, Tokyo Stock Company Ricoh Company (72) Inventor Yoshifumi Sakuramata 1-3-6 Nakamagome, Ota-ku, Tokyo Stock Company Ricoh Company (72) Inventor Junichi Takami 1-3-6 Nakamagome, Ota-ku, Tokyo Inside Ricoh Co., Ltd. (72) Inventor Ryo Akira 1-3-6 Nakamagome, Ota-ku, Tokyo F-term in Ricoh Co., Ltd. 5B064 AA10 FA16

Claims (7)

[Claims] 1. A character code / image separating means for separating a character code and an image character in a text, and a character recognition / character coding means for recognizing the separated image character and replacing the recognized character with a character code. Reading order control means for arranging the character-coded image characters and the original character codes in accordance with the character arrangement of the text,
A text-to-speech apparatus, comprising: voice synthesis means for voice-synthesizing an order-controlled character code; and converting an image character into a character code and reading it out. 2. The apparatus according to claim 1, further comprising: a visual importance determining unit configured to determine a visual attention level of the separated image characters, wherein the reading order is controlled according to a determination result of the visual importance determining unit. The text-to-speech apparatus according to claim 1, wherein the text includes image characters. 3. A visual importance determining means for determining a visual attention level of the separated image character, and controlling the speech synthesis unit in accordance with a determination result of the visual importance determining means. 2. The text-to-speech apparatus according to claim 1, wherein the auditory attribute of the separated image character portion is changed. 4. The text-to-speech apparatus according to claim 2, wherein a size of the image character is measured as the visual attention level. 5. The text-to-speech apparatus according to claim 2, wherein a contrast between a color of an image character and a background color is measured as the degree of visual attention. 6. The text-to-speech apparatus according to claim 2, wherein a shape characteristic of a character is measured as the visual attention level. 7. A single image, if coordinate information for dividing the image is set, is treated as independent character information for a plurality of corresponding areas. A text-to-speech device that includes image characters according to any one of the above.