JP4634461B2 - Document text-to-speech processing program and document browsing device - Google Patents

Description

  In the present invention, a structured document described in a markup language such as HTML (HyperText Markup Language) is displayed and processed by a user, and a text element of the displayed document is processed by speech synthesis and output. The present invention relates to an in-document text-to-speech processing program for causing a computer to execute text-to-speech processing, and a document browsing apparatus having processing means for executing the in-document text to-speech processing.

  As one of the data processing functions of a computer, there is a text-to-speech processing function that outputs text data after speech synthesis processing. With this text-to-speech processing function, the user can grasp the contents by listening to the audio data output from the speaker instead of browsing the document output by the display device or the printing device.

  For example, in an HTML document browsing program for displaying a web page (Web page) that is information on the Internet, speech data generated by speech synthesis processing of a text data portion in the web page displayed on the display device is output. Some have a text-to-speech processing function.

  However, web pages written in HTML are created on the assumption that they are displayed on the display device and allowed to be viewed by the user, so they are specified by elements such as navigation menus for viewing operations, advertisement banners, and anchor tags. The link information etc. which were done are incorporated. Since such navigation menus and advertisement banners may contain character strings, if all character strings in an HTML document are treated as the target of reading-out processing, various information is mixedly output. Thus, the user is also asked about unnecessary information.

Then, in order to exclude the information for the browsing operation from the target of the reading process, the navigation menu portion in the web page is changed using the DOM (Document Object Model) tree information generated for the web page display process. A method is known that discriminates and excludes text data of these portions from the target of the reading process and executes the reading process (see, for example, Patent Document 1).
JP 2004-171111 A

  Some Internet web pages have a series of pages configured according to a predetermined format so that the user can easily browse. For example, some web pages of news sites provided by a news server commonly use a configuration in which a header part, a main part, and a footer part are provided on each page. Each page displays the link information to the related page for navigation in the header, the supplementary explanation and copyright of the page and service in the footer, and the news that is the original provision purpose in the main part I try to let them.

  When reading out multiple pages configured in such a unified format, the same contents are displayed in the header and footer sections. Is duplicated for each page, which is troublesome for the user.

  Furthermore, when there is link information in the text data being read out, if the link destination information has already been read out, there is no need to read the information again by automatic jump processing.

  An object of the present invention is to efficiently read out information required by a user by removing information that is repeatedly displayed between web pages or information that has been read out from the target of the reading process and performing the reading process. It is to provide a program for causing a computer to execute a text-to-speech process that can be performed.

  Another object of the present invention is to provide a document browsing apparatus provided with processing means for executing the text reading process.

  According to the present invention, as a text-to-speech process in a document described and structured in a markup language on a computer, 1) a page for storing page configuration information of a document whose text elements have been read-out in page configuration information storage means Configuration information storage processing, 2) page configuration information acquisition processing for acquiring page configuration information of the document currently displayed, and 3) page configuration information and page configuration information storage means for the display target document A reading range in which the text elements that do not match the text elements of the stored page structure information are extracted from the page structure information of the document to be displayed, and the extracted text elements are set as the reading target. Setting process, and 4) display processing target document And reading processing and outputs the conversion process to the audio data set text elements processed speech Te, a program for executing.

  The present invention operates as follows.

  In the computer in which the present invention is installed, for example, a page structure for HTML document display processing such as DOM tree information of an HTML document in which a text element is read out, a document that is analyzed for each element and the hierarchical structure of each element. The page configuration information is stored in the page configuration information storage means. Then, the page configuration information of the document to be displayed is compared with the page configuration information of the previously read-out document stored in the page configuration information storage means, and the page of the document to be displayed is compared. Text elements other than text elements that overlap in the configuration information and the stored page configuration information are extracted, and the extracted text elements are set as a reading process target. Then, a reading process is performed in which the text element set as the reading process target in the document currently being displayed is converted into voice data and output.

  When a computer reads out a plurality of pages with the same page layout, the computer reads out the HTML document by excluding the text elements that are displayed in common on the plurality of pages from the target of the reading processing. Can do. Thus, the user is freed from the trouble of repeatedly reading out the same content, and can efficiently listen to necessary information.

  Further, according to the present invention, when link information is detected from a text element in 5) reading processing, the present invention acquires page configuration information of a linked document set in the linked information, and links to the linked page. Comparing the configuration information with the page configuration information stored in the page configuration information storage means, and if the page configuration information at the link destination completely matches the page configuration information that has been read out, automatic jump processing with the link information is performed. A program for executing automatic jump processing to be suppressed.

  Alternatively, according to the present invention, 6) the hash function value is calculated from the page configuration information of the document using a predetermined hash function, and the hash function value calculated from the page configuration information of the read-out document is stored in the computer. When the link information is detected from the text element in the hash data management process stored in the storage means and 7) the reading process, the hash function value of the page configuration information of the linked document set in the link information is acquired. Compare the hash function value of the linked document with the hash function value stored in the hash information storage means, and the hash function value of the linked document is exactly the same as the hash function value stored in the hash information storage means When doing this, an automatic job that suppresses automatic jump processing in link information A pump process, a program for executing.

  Alternatively, according to the present invention, 8) a document temporary storage process for temporarily storing a display-processed document in the document temporary storage means, and 9) a document stored in the document temporary storage means is read out to the computer. And a reading flag management process for setting a reading flag in the read-out document, and 10) in the reading-out process, when link information is detected from the text element, the link destination document set in the link information is To execute automatic jump processing that suppresses automatic jump processing with link information when a reading flag is set for a document stored in the document temporary storage means and stored in the document temporary storage means It is a program.

  Accordingly, the present invention determines whether or not the linked document has already been the target of the reading process even if the link information is set in the text element being read out, and has already been the target of the reading process. If so, automatic jump processing is not performed with the link information. Therefore, the user is freed from the annoyance that the content that has already been read out is repeatedly read out, and can listen to necessary information efficiently.

  In addition, the present invention is a document browsing apparatus having an in-document text reading processing function including processing means for executing processes realized by the in-document text reading processing program.

  According to the present invention, when a web page that uses the same layout as provided by a news site or the like is browsed and the content is read out, it is displayed in common in the header part and footer part of each page. The contents can be excluded from the target of the reading process. Therefore, the user is freed from the trouble of reading the same information repeatedly, so that only necessary information can be obtained efficiently and comfortably.

  Further, according to the present invention, even when link information is included in the text being read out, automatic jump processing can be suppressed if the link destination has already been read out. Thus, the user can efficiently obtain necessary information without reading the same page again and without wasting time.

FIG. 1 is a diagram showing a configuration example in the best embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a web page provided from a news site. FIG. 3 is a diagram illustrating an example of a DOM tree management list. FIG. 4 is a diagram illustrating an example of DOM tree information. FIG. 5 is a diagram showing a processing flow of the reading process in the first embodiment. FIG. 6 is a diagram showing a processing flow of the reading process in the first embodiment. FIG. 7 is a diagram showing a processing flow of the reading process in the first embodiment. FIG. 8 is a diagram showing a process flow of the URL list acquisition process in step S111. FIG. 9 is a diagram showing a processing flow of the automatic jump processing in step S8. FIG. 10 is a diagram illustrating an example of a hash management list. FIG. 11 is a diagram showing a processing flow of the reading process in the second embodiment. FIG. 12 is a diagram showing a process flow of the automatic jump process in step S20. FIG. 13 is a diagram illustrating an example of a cache management list. FIG. 14 is a diagram showing a processing flow of the reading process in the third embodiment. FIG. 15 is a diagram showing a process flow of the automatic jump process in step S30.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Document browsing apparatus 110 Browsing process control part 111 Communication process part 112 HTML analysis process part 113 Layout process part 114 Screen display process part 115 Temporary storage document management part 116 Reading flag management part 117 Document temporary storage part 120 Reading process control part 121 DOM Tree saving processing unit 122 DOM tree storage unit 123 Reading processing unit 124 Reading range determination unit 125 Automatic jump processing unit 126 Hash data management unit 141 Speech synthesis processing unit 143 Audio output processing unit 2 Display device 3 Speaker

  FIG. 1 shows a configuration example in the best embodiment of the present invention. In the present embodiment, an example of a web page (HTML document) provided on the Internet will be described as an example of a document described and structured in a markup language.

  The document browsing apparatus 1 is a computer for browsing a document having a reading processing function for text in a document. In this example, the in-document text reading processing program of the present invention is installed in the memory in the document browsing apparatus 1 and is executed at a predetermined opportunity.

  The document browsing apparatus 1 includes a display device 2, a speaker 3, a browsing processing control unit 110, a communication processing unit 111, an HTML analysis processing unit 112, a layout processing unit 113, a screen display processing unit 114, a temporarily stored document management unit 115, and a reading flag. Management unit 116, document temporary storage unit 117, reading processing control unit 120, DOM tree saving processing unit 121, DOM tree storage unit 122, reading processing unit 123, reading range determination unit 124, automatic jump processing unit 125, hash data management unit 126, a speech synthesis processing unit 141, and a speech output processing unit 143.

  The browsing process control unit 110 is a processing unit that acquires a web page and controls a series of processes related to the browsing process that is displayed on the display device 2. In addition, when the user requests a reading process, the browsing process control unit 110 notifies the reading process control unit 120 of the reading process request.

  The communication processing unit 111 is a processing unit that acquires a web page from a website server according to the HTTP protocol.

  The HTML analysis processing unit 112 is a processing unit that analyzes the HTML tag of the web page and generates page configuration information converted into a format called a DOM tree.

  The layout processing unit 113 is a processing unit that lays out elements of a web page based on page configuration information (hereinafter referred to as DOM tree information).

  The screen display processing unit 114 is a processing unit that displays individual elements of the layout-processed web page on the display device 2.

  The temporary storage document management unit 115 is processing means for temporarily storing the web page (HTML document) displayed on the display device 2 by the browsing processing control unit 110 in the document temporary storage unit 117.

  When the web page stored in the document temporary storage unit 117 is read out by the read-out processing unit 123, the read-out flag management unit 116 is a processing unit that manages read-out information in which a read-out flag is set for the web page.

  The document temporary storage unit 117 is a unit that temporarily stores the web page displayed on the display device 2 by the browsing processing control unit 110.

  The reading process control unit 120 is a processing unit that receives a processing request from the user and controls a series of processes related to the reading process in which a predetermined text element of the web page displayed on the display device 2 is subjected to speech synthesis processing and output. .

  The DOM tree save processing unit 121 is a processing unit that stores the DOM tree information of the web page that is the target of the reading process in the DOM tree storage unit 122.

  The reading processing unit 123 is a processing unit that sequentially acquires the text elements of the web pages to be processed by the reading range determination unit 124 from the top, and executes the reading processing by the voice synthesis processing unit 141 and the voice output processing unit 143. is there.

  The reading range determination unit 124 compares the DOM tree information of the web page to be displayed with the DOM tree information of the already read-out web page stored in the DOM tree storage unit 122, and displays the DOM tree to be displayed. This is a processing means for extracting a text element that does not match the text element of the stored DOM tree information from the text elements of the information and setting the extracted text element as a processing target (reading processing target) of the reading processing unit 123.

  When the automatic jump processing unit 125 detects link information from a text element set as a reading process target, the automatic jump processing unit 125 acquires the DOM tree information of the web page set as the link destination of the link information, and the DOM tree information of the link destination Is compared with the DOM tree information stored in the DOM tree storage unit 122 by the saving process. When the link destination DOM tree information completely matches the DOM tree information in the DOM tree storage unit 122, the link information is automatically When the jump process is suppressed (invalidated) and the DOM tree information of the link destination does not completely match the DOM tree information of the DOM tree storage unit 122, the automatic jump process is performed based on the link information, and the link destination It is a processing means for acquiring a web page (HTML document).

  Further, when link information is detected from the text element that is the processing target of the reading processing unit 123, the automatic jump processing unit 125 generates a hash function value (hash) generated from the linked web page set in the link information. Data) and the hash data generated from the read-out processed web page saved in the hash data management unit 126, and the hash data of the linked web page is saved in the hash data management unit 126 When it completely matches the hash data of the page, the automatic jump processing with the link information is suppressed.

  Further, when link information is detected from the text element that is the processing target of the reading processing unit 123, the automatic jump processing unit 125 caches the link destination web page set in the link information in the document temporary storage unit 117. Further, based on the reading information managed by the reading flag management unit 116, a value indicating that reading has been performed is set in the reading flag of the linked web page cached in the temporary document storage unit 117. If so, the automatic jump processing with the link information is suppressed.

  The hash data management unit 126 creates hash data using a predetermined hash function based on the web page that has been read out, and the correspondence between the storage location information (such as URL) of the web page and the hash data of the web page Is a processing means for generating and managing a cache management list indicating

  In addition, the hash data management unit 126 generates hash data based on the linked web page passed from the automatic jump processing unit 125, and sends the generated hash data of the linked web page to the automatic jump processing unit 125. return.

  The speech synthesis processing unit 141 is a processing unit that converts the text element to be processed acquired from the reading processing unit 123 into a phonetic character string and converts the phonetic character string into waveform data. The audio output processing unit 143 is processing means for outputting the waveform data as audio from the speaker 3.

  Next, a specific processing example of the present invention will be described.

  The document browsing device 1 displays a web page provided from a news site as shown in FIG. 2 on the display device 2 so as to allow the user to browse, and a text element (text data) in the web page according to a user request. Is output from the speaker 3. Here, it is assumed that all the web pages provided on the news site are created according to a specific layout so that the user can easily browse.

  FIG. 2A is a diagram showing an example of a web page A having a news headline as content (hereinafter referred to as content A), and FIG. 2B is a content of a news body corresponding to a news headline (hereinafter referred to as content). It is a figure which shows the example of the web page B set to B).

  The web page A and the web page B are composed of a header part H, a footer part F, and a main part Main according to a common layout, and the same contents are displayed on the header part H and the footer part F. Moreover, the web page B is set as one link destination set to the news headline (for example, news headline HL1) displayed on the main part A_Main of the web page A. In the main part B_Main of the web page B, a news body NEWS1 corresponding to the news headline HL1 of the link source is displayed.

[First Example]
In the first embodiment, the document browsing apparatus 1 determines whether or not automatic jump processing is necessary for a range of text elements to be read out and link information using DOM tree information of a web page.

  In the document browsing apparatus 1 of this example, the hash data management unit 126, the temporary storage document management unit 115, the reading flag management unit 116, and the document temporary storage unit 117 shown in the configuration example of FIG. Absent.

  In the document browsing apparatus 1, when the communication processing unit 111 downloads the web page A from the news site, the HTML analysis processing unit 112 analyzes the HTML tag of the downloaded web page A and generates DOM tree information A. . Then, the layout processing unit 113 lays out each element of the web page A based on the DOM tree information A, and the screen display processing unit 114 displays the web page A subjected to the layout process on the display device 2.

  Further, the reading processing unit 123 acquires the DOM tree information A of the web page A through the reading processing control unit 120. Based on the DOM tree management list, the reading range determination unit 124 extracts the DOM tree information of web pages belonging to the same domain as the web page A one by one from the DOM tree storage unit 122, and extracts the DOM tree information A and the extracted DOM. The tree information is compared from the beginning and end, and it is determined whether or not the same text element exists in both.

  The DOM tree management list is a list for managing DOM tree information of web pages that have been read out and saved in the DOM tree storage unit 122. FIG. 3 shows an example of the DOM tree management list. The DOM tree management list includes web page storage location information (URL) and DOM tree addresses that are pointer information to DOM tree information in the DOM tree storage unit 122.

  Here, the web page A is the first web page downloaded from the news site, there is no text element that matches the DOM tree information of the web page that has been read out and saved in the DOM tree storage unit 122, and the web page A Assume that all text elements in are not yet read out.

  The reading range determination unit 124 determines that all text elements of the DOM tree information A are the range of the reading process, and sets the first text element to the last text element of the DOM tree information A as a reading process target. For example, “title, latest news, politics, economy, society, overseas, news headline HL1...” On the web page A in FIG.

  The reading processing unit 123 sequentially passes the text elements of the DOM tree information A of the web page A to the speech synthesis processing unit 141. The speech synthesis processing unit 141 converts the text element to be read out into a phonetic character string, further converts the phonetic character string into waveform data, and the voice output processing unit 143 uses the waveform data as a voice as a speaker. 3 is output.

  Then, the DOM tree save processing unit 121 stores the DOM tree information A of the web page A that has been read out in the DOM tree storage unit 122, and stores the URL of the web page A and the DOM tree information A in the DOM tree management list. Pointer information (DOM tree address) is added.

  Furthermore, it is assumed that link information that links the web page B to the text element “news headline HL1” of the web page A to be read out is set.

  The automatic jump processing unit 125 acquires the DOM tree information B of the web page B that is the link destination of the link information “news headline HL1” via the reading processing control unit 120. Then, the acquired DOM tree information B is compared with the DOM tree information stored in the DOM tree storage unit 122, and it is determined whether there is a completely matching DOM tree information.

  Here, since the web page B has not yet been subjected to the reading process, the DOM tree storage unit 122 does not store anything that completely matches the DOM tree information B.

  Since the automatic jump processing unit 125 has not detected the completely matched DOM tree information, the automatic jump processing unit 125 executes automatic jump processing on the link information “news headline HL1”. By this automatic jump processing, the web page B in FIG. 2B is downloaded as a display processing target.

  The reading range determination unit 124 extracts and compares the DOM tree information B of the web page B and the DOM tree information of the read web page stored in the DOM tree storage unit 122 one by one. Determine if there is.

  FIG. 4 shows an example of DOM tree information. 4A shows an example of the DOM tree information A of the web page A (content A), which is one of the DOM tree information stored in the DOM tree storage unit 122. FIG. 4B shows the display process. It is a figure which shows the example of the DOM tree information B of the web page B (contents B) used as the object of.

  The reading range determination unit 124 compares the text elements of the DOM tree information B and the DOM tree information A stored in the DOM tree storage unit 122 in order from the beginning and the end. In the comparison process, some text elements “latest news, politics, economy, society, overseas” from the top of the DOM tree information B and some text elements “supplementary explanation, Copyright (C)... It is detected that the text element of the DOM tree information A is duplicated. Further, it is assumed that the text element “news text” of the DOM tree information B does not match the text elements of the DOM tree information A and other DOM tree information.

  The reading range determination unit 124 sets, from the DOM tree information B of the web page B, a text element “news body” (for example, news body NEWS1) other than the overlapping text elements as a reading process target. As in the case of the web page A, the reading processing unit 123 passes the text element to be read out to the speech synthesis processing unit 141 and performs the reading processing.

  Thereby, in the reading process of the web page B, the part that has already been read out in the reading process of the web page A is not read out, and the user can avoid the situation that the same content is read out.

  Thereafter, when another link information is detected from the text element of the web page A, the automatic jump processing unit 125 performs the automatic jump processing based on the DOM tree information of the link destination of the link information as in the above processing. Determine whether or not.

  When the reading range determination unit 124 acquires the DOM tree information C of another web page C that is the next display processing target, the range of the text element that is the target of the reading processing is the same as the processing described above. Determine.

  5 to 7 show a processing flow of the reading process in the first embodiment.

  In the processing flow of FIG. 5, the reading processing unit 123 turns off the initial value of the reading processing flag (hereinafter referred to as processing flag) indicating whether or not to execute the reading processing (step S10). The reading range determination unit 124 executes a reading range determination process (step S11).

  6 and 7 show a processing flow of the reading range determination process.

  The reading range determination unit 124 stores the element number of the first text element of the DOM tree information X for the current display processing target web page X in NODE_S and WORK_S, and the same DOM tree information X in NODE_E and WORK_E. The element number of the last text element is stored (step S110). Here, NODE_S is a storage area for storing the element number of the text element at which the reading process is started, and NODE_E is the element number of the text element at which the reading process is ended.

  Next, in order to acquire the DOM tree information of the web page belonging to the same domain from the DOM tree storage unit 122, a URL list acquisition process from the DOM tree management list is executed (step S111). Details of the URL list acquisition process will be described later.

  Then, while URLs that have not been subjected to the determination process remain in the URL list extracted by the URL list acquisition process (YES in step S112), the processes in and after step S113 are performed, and if there is no URL left unprocessed. (NO in step S112), the reading range determination process is terminated.

  First, when a URL that has not been subjected to determination processing remains in the URL list (YES in step S112), text elements are extracted in order from the top of the DOM tree information X that is the current display processing target (step S113). Further, text elements are extracted in order from the top of one DOM tree information Y extracted from the DOM tree information under the same domain saved in the DOM tree storage unit 122 (step S114). If the text element extracted in step S113 and the text element extracted in step S114 are the same text element (YES in step S115), the element number of the text element currently extracted in DOM tree information X is stored in WORK_S. (Step S116).

  If the extracted text elements are not the same (NO in step S115), the text elements are extracted in order from the end of the currently displayed DOM tree information X (step S117). Further, the text elements are extracted in order from the end of the saved DOM tree information Y under the same domain (step S118). If the text element extracted in step S117 and the text element extracted in step S118 are the same text element (YES in step S119), the element number of the text element currently extracted in DOM tree information X is stored in WORK_E. (Step S120).

  If the extracted text elements are not the same (NO in step S119), “the element number of the last text element of the current DOM tree information X is stored in WORK_S, and the current DOM tree is stored in WORK_E. When the element number of the first text element of information X is stored "(FIG. 7: YES in step S121), all text elements are duplicated, so the values of WORK_S and WORK_E are cleared. That is, a state without a value is set (step S122).

  Also, “the element number of the last text element of the current DOM tree information X is stored in WORK_S, and the element number of the first text element of the current DOM tree information X is stored in WORK_E”. If not (NO in step S121), the range from the element number of WORK_S to the element number of WORK_E is compared with the range from the element number of NODE_S to the element number of NODE_E (step S123).

  If the range from the element number of WORK_S to the element number of WORK_E is within the range from the element number of NODE_S to the element number of NODE_E (YES in step S123), the element number +1 of WORK_S is stored in NODE_S, and WORK_E Element number-1 is stored in NODE_E (step S124).

  When the reading range determination process is completed, the reading processing unit 123 sequentially extracts text elements from the current DOM tree information X set in the reading range determination process (FIG. 5: Step S12), and the processing flag = ON. If so (YES in step S13), the text of the extracted text element is read out (step S14). On the other hand, if the processing flag is not ON (NO in step S13), it is determined whether or not the extracted text element is a text element for starting the reading process (step S15). That is, if the element number of the extracted text element is NODE_S, it is determined that the extracted text element is a text element for starting the reading process (YES in step S15), and the processing flag is set to ON ( In step S16), the text element is read out (step S14). If the element number of the extracted text element is not the element number of NODE_S, it is determined that the extracted text element is not a text element for starting the reading process (NO in step S15), and the process returns to step S12.

  Thereafter, when link information is detected from the text element being read out (YES in step S17), the automatic jump processing unit 125 executes automatic jump processing (step S18). The automatic jump process will be described later.

  If link information is not detected from the text element being read out (NO in step S17), it is determined whether the text element being read out is a text element that has been read out (step S19). . That is, if the element number of the extracted text element is an element number of NODE_E, it is determined that the extracted text element is a text element at the end of the reading process (YES in step S19), and the reading process is terminated.

  If the element number of the extracted text element is not the element number of NODE_E, it is determined that the extracted text element is not a text element for which the reading process has been completed (NO in step S19), and the process returns to step S12.

  FIG. 8 shows a process flow of the URL list acquisition process in step S111 of FIG.

  The reading range determination unit 124 stores the URL of the web page that is the target of the current display process in URL_A (step S1110), and the partial character string from “http: //” to the next “/” of URL_A. Is stored in DOMAIN_A (step S1111).

  Then, while DOM tree information that has not been subjected to comparison processing remains in the DOM management list (YES in step S1112), the loop processing of steps S1113 to S1116 is performed (step S1117).

  First, URLs of DOM tree information saved in the DOM tree storage unit 122 are sequentially acquired from the DOM tree management list and stored in URL_B (step S1113). Then, the partial character string from “http: //” to the next “/” of URL_B is stored in DOMAIN_B (step S1114). DOMAIN_A and DOMAIN_B are compared (step S1115), and if DOMAIN_A and DOMAIN_B completely match (YES in step S1115), the URL stored in URL_B is stored in the URL list (step S1116).

  If there is no DOM tree information to be compared in the DOM management list by the loop process (step S1117), the process ends.

  FIG. 9 shows a process flow of the automatic jump process in step S18 of FIG.

  The automatic jump processing unit 125 saves the URL and DOM tree information of the current display target web page in the temporary storage area (step S180), and the web page set as the link destination by the browsing processing control unit 110. Is acquired (step S181), the DOM tree information of the acquired link destination web page is acquired through the reading process control unit 120 (step S182). Then, a reading range determination process is performed (step S183). In this reading range determination process, the same processing as the reading range determination process in step S11 (FIG. 5) is performed.

  Then, in the reading range determination process, it is determined whether or not the text element as the range of the reading process has been extracted from the DOM tree information of the linked web page (step S184). If the text element that is the range of the reading process can be extracted from the DOM tree information of the linked web page (YES in step S184), the browsing process control unit 110 displays the linked web page (step S185). ), The reading processing unit 123 performs the reading processing of the text element determined as the reading range (step S186). This reading process is the same as the reading process shown in FIG.

  On the other hand, the same DOM tree information as the DOM tree information of the linked web page is stored in the DOM tree storage unit 122, and the text element as the range of the reading process can be extracted from the DOM tree information of the linked web page. If not (NO in step S184), the process is terminated.

[Second Embodiment]
The document browsing apparatus 1 in the second embodiment specifies the range of text elements to be read out using the DOM tree information of the web page, and determines whether or not the automatic jump process is required in the link information. Judgment is made using hash data generated from the information.

  In the document browsing apparatus 1 of this example, the temporarily stored document management unit 115, the reading flag management unit 116, and the document temporary storage unit 117 shown in the configuration example of FIG. 1 are not essential components.

  The reading range determination unit 124 of the document browsing apparatus 1 specifies the range of the reading processing by the same processing as in the first embodiment. The DOM tree save processing unit 121 then passes the DOM tree information of the web page read-out by the read-out processing unit 123 to the hash data management unit 126.

  The hash data management unit 126 generates hash data (hash character string) from the DOM tree information using a predetermined hash function, and manages the hash data (hash character string) using the hash management list.

  The hash management list is a list for managing correspondence with hash data generated from DOM tree information of a web page. FIG. 10 shows an example of the hash management list. The hash management list stores web page storage location information (URL) and hash data (hash function value) generated from the DOM tree information of the web page.

  When link information is detected from the reading process target, the automatic jump processing unit 125 acquires the DOM tree information of the web page set as the link destination, passes it to the hash data management unit 126, and the link destination. The hash data generated from the DOM tree information of the web page is acquired. Furthermore, when the hash data of the link destination web page is compared with the hash data of the hash management list and hash data that exactly matches the link destination hash data is searched from the hash management list, the link information is automatically Disable jump processing.

  In this way, by using hash data to determine whether or not the linked web page is already subject to read-out processing, it becomes possible to perform processing for determining the necessity of automatic jump processing at a higher speed. .

  FIG. 11 shows a processing flow of the reading process in the second embodiment.

  In the processing flow shown in FIG. 11, the processing steps to which the same numbers as the processing steps of the reading processing flow of the first embodiment shown in FIG. 5 are assigned the same processing as the processing steps of the processing flow of FIG. Means to do.

  In the processing flow of FIG. 11, the automatic jump processing unit 125 performs another automatic jump processing instead of the processing of step S18 shown in FIG. 5 (step S20).

  Further, after the processing of step S19, the hash data management unit 126 generates hash data from the read-out DOM tree information, and hashes the corresponding web page URL and the hash data generated from the DOM tree information of the web page. Store in the management list (step S21).

  FIG. 12 shows a process flow of the automatic jump process in step S20 of FIG.

  The automatic jump processing unit 125 saves the URL and DOM tree information of the current display target web page in the temporary storage area (step S200), and acquires the DOM tree information of the web page set as the link destination. (Step S201), hash data is generated from the acquired DOM tree information (Step S202), and the generated hash data is stored in HASH_A (Step S203).

  Then, while the hash data to be compared remains in the hash management list (YES in step S204), the loop processing in steps S205 to S206 is performed (step S207).

  First, hash data is sequentially extracted from the hash management list and stored in HASH_B (step S205). The hash data (hash character strings) stored in HASH_A and HASH_B are compared (step S206). If HASH_A and HASH_B do not completely match and there is no remaining in the hash management list (NO in step S206, In step S207, the browsing processing control unit 110 displays the linked web page (step S208), and the text of the web page is read out (step S209). This reading process is the same as the reading process shown in FIG.

  If HASH_A and HASH_B are completely matched (YES in step S206), the linked web page has already been read out, and the process is terminated.

[Third embodiment]
The document browsing apparatus 1 according to the third embodiment determines the necessity of automatic jump processing in the range of text elements to be read out and link information using the DOM tree information of the web page.

  In the document browsing apparatus 1 of this example, the hash data management unit 126 shown in the configuration example of FIG. 1 is not an essential component.

  When the communication processing unit 111 acquires a web page to be displayed during document display processing by the browsing processing unit 110, the temporary storage document management unit 115 of the document browsing apparatus 1 stores the web page as a document temporary storage unit 117. Store temporarily.

  Then, when the text of the web page stored in the document temporary storage unit 117 is the target of the reading process, the reading flag management unit 116 sets a value indicating that the reading process has been completed to the reading flag of the web page. The URL and the reading flag of the web page cached in the document temporary storage unit 117 are managed using the cache management list.

  FIG. 13 shows an example of the cache management list. The cache management list is a list for managing web page storage location information (URL), a cache file name in the document temporary storage unit 117 of the web page, a reading flag, and the like.

  The reading flag is set to ON for a web page that has been read out, and OFF for a web page that has not been read out.

  When link information is detected from the range of the reading process, the automatic jump processing unit 125 acquires the URL set as the link destination of the link information, refers to the cache management list, and determines the link destination URL. When it is possible to search for a web page whose URL matches completely and whose reading flag is ON, the automatic jump processing with this link information is invalidated.

  In this way, the automatic jump is performed by using the reading flag that is given to the web page temporarily stored for display processing to determine whether or not the linked web page is already subject to reading processing. Processing for determining whether processing is necessary can be performed at higher speed.

  FIG. 14 shows a processing flow of the reading process in the third embodiment.

  In the processing flow shown in FIG. 14, the processing steps to which the same numbers as the processing steps of the reading processing flow of the first embodiment shown in FIG. 5 are assigned the same processing as the processing steps of the processing flow of FIG. Means to do.

  In the processing flow of FIG. 14, the automatic jump processing unit 125 performs another automatic jump processing instead of the processing of step S18 shown in FIG. 5 (step S30).

  Further, after the processing in step S19, the reading flag management unit 116 adds the URL of the web page cached in the document temporary storage unit 117 to the cache management list when the displayed web page is read out. The reading flag is set to ON (step S31).

  FIG. 15 shows a process flow of the automatic jump process in step S30 of FIG.

  The automatic jump processing unit 125 saves the URL of the current display target web page and the DOM tree information in the temporary storage area (step S300), and sets the URL of the web page set as the link destination from the cache management list. It is acquired (step S301), and it is determined whether or not there is a link destination URL in the cache management list (step S302). If there is no URL that completely matches the link destination URL in the cache management list (NO in step S302), it is further determined whether or not the reading flag of the corresponding URL in the cache management list is ON (step S303). Then, if the link destination URL reading flag is not ON (NO in step S303), the DOM tree information of the acquired link destination web page is acquired (step S304), and the browsing processing control unit 110 uses the link destination web. The page is displayed (step S305), and the text of the web page is read out (step S306). This reading process is the same as the reading process shown in FIG. In addition, when there is a URL that completely matches the link destination URL in the cache management list (YES in step S302), or when the reading flag of the corresponding URL in the cache management list is ON (YES in step S303). Terminates the process.

  Although the present invention has been described above with reference to the embodiment, it is obvious that the present invention can be variously modified within the scope of the gist thereof. Further, the present invention can be implemented as a processing program that is read and executed by a computer. The program for realizing the present invention can be stored in an appropriate recording medium such as a portable medium memory, a semiconductor memory, and a hard disk, which can be read by a computer, provided by being recorded on these recording media, or communication. It is provided by transmission / reception using various communication networks via an interface.

Claims (5)

As a text-to-speech process in a document written and structured in markup language on a computer
Page configuration information storage processing for storing page configuration information of a document whose text element has been read out in the page configuration information storage means;
A page configuration information acquisition process for acquiring the page configuration information of the document currently displayed;
The page configuration information of the document to be displayed is compared with the page configuration information stored in the page configuration information storage means, and from the page configuration information of the document to be displayed, the text element of the stored page configuration information A text range that does not match the text element, and sets the read text element as a target for the reading process;
Read-out processing for converting the text element set as the read-out processing target into speech data in the display-processing target document and outputting the voice data,
Text-to-speech processing program for document to be executed. In the computer,
When link information is detected from the text element in the reading process, the page configuration information of the link destination document set in the link information is acquired, and the link destination page configuration information and the page configuration information storage are acquired. The page configuration information stored in the means is compared, and when the page configuration information at the link destination completely matches the page configuration information that has been read out, automatic jump processing that suppresses automatic jump processing with the link information Processing
The in-document text-to-speech processing program according to claim 1 for execution. In the computer,
A hash data management process for calculating a hash function value using a predetermined hash function from the page configuration information of the document and storing the hash function value calculated from the page configuration information of the read-out document in a hash information storage unit;
When link information is detected from the text element in the reading process, a hash function value of the page configuration information of the link destination document set in the link information is acquired, and a hash function value of the link destination document And the hash function value stored in the hash information storage means, and when the hash function value of the linked document completely matches the hash function value stored in the hash information storage means, the link Automatic jump processing that suppresses automatic jump processing with information,
The in-document text-to-speech processing program according to claim 1 for execution. In the computer,
A temporary document storage process for temporarily storing the displayed document in the temporary document storage means;
A reading flag management process for setting a reading flag for the read-out document when the document stored in the document temporary storage means is read-out;
In the reading process, when link information is detected from the text element, the link destination document set in the link information is stored in the document temporary storage means and stored in the document temporary storage means. Automatic jump processing for suppressing automatic jump processing in the link information when a reading flag is set in the document,
The in-document text-to-speech processing program according to claim 1 for execution. In a document browsing apparatus that displays and processes a structured document described in a markup language,
Page configuration information storage means for storing page configuration information of a document whose text element has been read out;
Page configuration information storage means for storing page configuration information of the display-processed and read-out document in the page configuration information storage means;
A page configuration information acquisition means for acquiring page configuration information of a document currently displayed;
The page configuration information of the document to be displayed is compared with the page configuration information stored in the page configuration information storage means, and from the page configuration information of the document to be displayed, the text element of the stored page configuration information A text range that does not match the text element, and sets the read text element as the text processing target,
A document browsing apparatus comprising: a reading processing unit that converts a text element set as the reading processing target in the display processing target document into voice data and outputs the voice data.

Images