Abstract: The present invention relates to creating a web page and voice browsing of the web page, and more particularly, it improves accessibility for the voice browsing of the web page through a synthetic voice, efficiently with high reliability. A content creation system 20 of the present invention is used for creating a content which may be viewed through the synthetic voice, the system including: a database 22 for storing a structured document; and an information process section 24 for creating a speech node series 18 from the structured document, and calculating a reaching time from starting voice synthesis of the speech node series 18 until each node is outputted as the synthetic voice. The information process section 24 includes a support process section 36 to determine a graphic display corresponding to the reaching time, and to visually display the reaching time to a predetermined node by the voice synthesis on a screen of a display section 26.

Abstract: Provides evaluating programs that enable problems to be identified in a manner that reflects actual conditions for an evaluation of contents of interest. In an example embodiment, there is provided a program for evaluating contents of interest that causes a computer to implement the functions of: performing a primary evaluation of contents of interest on the basis of one evaluation criterion; performing a primary evaluation of the contents of interest on the basis of another evaluation criterion; and performing a secondary evaluation of the contents of interest on the basis of a plurality of the first evaluations.

Abstract: The present invention provides effective information search means, and/or effective acquired information submission means, without overtly expressing an intent (e.g., through the depression of a search button) to acquire information. In an example embodiment, the kana-kanji conversion routine is activated, and a character string is input using voice, a keyboard or a graphic entry process. Then, a conversion key is depressed to convert the input character string into kanji. Upon the depression of the conversion key, the homonym candidate selection routine is initiated, and the conversion candidate is presented. In response to the depression of the conversion key, or the change of the conversion candidate in the homonym candidate selection routine, the information access routine is activated. Then, the information access procedure is performed, and the search results are acquired. Thereafter, the search results are presented.

Abstract: A device, computer program and method for outputting linguistic information. The voice output device, for example, includes an output information acquisition unit acquiring linguistic information and attribute information. Attribute information includes an attribute added to each linguistic element included in the linguistic information. A tactile pattern storage unit stores a predetermined tactile pattern corresponding to each linguistic element. A tactile pattern acquisition unit acquires the tactile pattern from the tactile pattern storage unit. A voice output unit reads aloud the linguistic elements and a tactile pattern output unit outputs, in parallel with reading aloud each linguistic element, the tactile pattern corresponding to the attribute added to the linguistic element, thereby allowing a user to sense the tactile pattern by the sense of touch.

Abstract: Target subtree setting means sets a target subtree relating to a content portion. Occurrence mode detecting means collates a target subtree relating to a content with a tree relating to each of past structured/hierarchical contents and detects an occurrence mode of each node of the target subtree. Statistical information generating means generates statistical information concerning an occurrence frequency of the occurrence mode of each node in the target subtree. Classifying means classifies each node of the target subtree based on the statistical information and a result of detecting the occurrence mode. Matching pattern generating means generates the matching pattern for the target content portion based on the classification. The structured/hierarchical contents are identified by use of the matching pattern.

Abstract: Provides ability to analyze the readability of an image to be displayed on a screen as a web page, etc., and appropriately modify the image. It includes a rendering section for generating an image by rendering an HTML document; an image processing section for performing image processing on the image generated by the rendering section to simulate and evaluate how the image is viewed under a certain visual characteristic; and a result presentation section for locating an HTML element that needs to be modified in the HTML document to be processed based on the evaluation result of the image processing section 30 and for presenting the HTML element to a web page creator. The result presentation section also retrieves a modification method for the HTML element that needs to be modified from a symptom model storage section. A document modification processing section actually modifies the HTML document.

Abstract: The present invention provides effective information search means, and/or effective acquired information submission means, without overtly expressing an intent (e.g., through the depression of a search button) to acquire information. In an example embodiment, the kana-kanji conversion routine is activated, and a character string is input using voice, a keyboard or a graphic entry process. Then, a conversion key is depressed to convert the input character string into kanji. Upon the depression of the conversion key, the homonym candidate selection routine is initiated, and the conversion candidate is presented. In response to the depression of the conversion key, or the change of the conversion candidate in the homonym candidate selection routine, the information access routine is activated. Then, the information access procedure is performed, and the search results are acquired. Thereafter, the search results are presented.

Abstract: Digest screen display content deciding means selects display elements belonging to respective regions of a document based on display priorities of the display elements, which are obtained by digest screen display priority information creating means, and decides selected display elements as display content of a digest screen under a condition where a total display area does not exceed a required display area. A merging relationship among the regions is set based on layout information for the regions, created by digest screen region layout information creating means. Display content deciding means decides the display content of a detail screen based on the merging relationship among the regions, and creates a digest of the detail screen based on control information created by control information creating means. Moreover, digest screen display content changing means changes the display content of the digest screen in response to an operation of a user.

Abstract: Method, system, and program product for analyzing a generation program for generating document information consisting of a plurality of document components. The the generation program contains commands for generating the document components, contained in the document information. The information processing system includes: a correspondence information generator for executing the generation program and generating correspondence information which indicates correspondence between the plurality of generating commands and document components generated by the generating commands; a selector for selecting any of the plurality of document components at the instruction of a user; and a command information output unit for selecting, based on the correspondence information, the generating command which has generated the document component selected by the selector and outputting information for identifying the generating command.

Abstract: The method, system, and program product enable a receiving-side user to speedily obtain important information in tree-structured documents even if the size of the received tree structure is large. This is accomplished by setting a node priority with respect to each of nodes of a tree-structured document. The priority is set based on the basis of the importance of an information portion presented from the node to a receiving-side user. Node stream generation means generate a node stream in which nodes and/or subtrees of a tree-structured document to be transmitted are arranged in sequence according to node priorities. Transmitting means send the node stream to tree-structured document receiving apparatus via a network. In tree-structured document receiving apparatus, extraction means extract the nodes and/or subtrees in the node stream according to the sequence of arrangement, and reconstruction means add the nodes and/or subtrees in the extraction order to the tree-structured document under reconstruction.

Abstract: To efficiently search an evaluation result of a plurality of XPath expressions with respect to a data file such as an XML document: an evaluation result of an XPath expression is obtained by generating a data structure with a redundant element by evaluating what common part or dependency has been omitted from a plurality of XPath expressions to be evaluated, and then the data structure is used with respect to a data file to be processed.

Abstract: A system and method for transcoding digital content (e.g. web content) by correctly employing one annotation for multiple digital contents. This can efficiently reduce the workloads required for the addition of annotation data during the transcoding process. A transcoding system comprises an annotation database system for storing annotation data to be used for the transcoding of contents, and a transcoder for transcoding the contents based on annotation data stored in the annotation database system. Upon receiving an inquiry from the transcoder, a correlation between elements in the contents and descriptions of the annotation data is checked to select one annotation that can be employed for transcoding the content. The correlation is specifically determined based on XPath information.

Abstract: To convert the structure of a Web page into contents that a user can easily listen to, and for permitting the user to access and obtain information without have to perform and navigation, in a manner similar to listening to the radio. A designated target page is obtained from a Web server, and a linked list for the target page is analyzed to obtain a linked destination page. A transcoding module inserts the main content of the linked destination page into a linked location on the target page, and converts the page into a structure appropriate for oral reading. Then, a VoiceXML generation module converts the target page into a VoiceXML document, and a VoiceXML browser performs a speech response process for the VoiceXML document. A command is input to the VoiceXML browser orally or by using a dial key entry provided using a telephone, and a speech from the VoiceXML browser is output to the telephone.

Abstract: The present invention provides improvement of operations providing annotation and layout for an HTML page file. In an example embodiment, a page acquisition module obtains page files from a web server, and an HTML file analysis module extracts tags and characteristic values related to the layout. A page group detection module employs layout tags and their characteristic values to group page files that have the same or a similar layout. When an annotation addition module adds an annotation to an arbitrary page file in the obtained layout group, the annotation is applied for another page file in the layout group. When the layout group is divided or layout groups are unified, a correction module for the function of distance calculation corrects a calculation expression for a distance between pages or layout groups in order to reflect the division or unification results obtained by the user.

Abstract: A technique for the simplification of Web pages in order to access necessary information rapidly, when displaying or outputting Web pages using a small screen device or a voice browser. A method for simplifying Web contents includes the steps of acquiring a target page subject to simplification, acquiring adjoining pages that adjoin the target page and performing a difference operation to delete objects that are common among the target page and the adjoining pages from the target page to generate a simplified page.