JP2002229843A - System and method for trans-coding information for user interface having restricted voice or display - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable an individual accessing a content of an internet through a device corresponding to audio use or mainly to the audio use accompanied by a visual form, to control his experience more intelligibly. SOLUTION: This method for trans-coding information provided in a first markup language comprises a step for selecting parts of information (524), determining complexity of the selected parts (526), preparing a hypertext link for accessing a subset of the information, and the subset of the information, when the complexity of the selected parts exceeds a limit (534), providing at least one of the selected parts with the hypertext link and the subset (538), and giving an annotation related thereto (542).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

This application is a continuation-in-part of U.S. Patent Application Serial No. 09 / 574,990, filed May 19, 2000, which is hereby incorporated by reference herein, and is a priority application therefrom. It asserts its rights. The present invention relates to a communication system for transcoding information created for a large visual display into information used in a small visual display, or information used in an audio format, such as in a telephone.

[0002]

BACKGROUND OF THE INVENTION Through the World Wide Web or the Internet, users access a variety of information, including, for example, weather forecasts, traffic information, movie listings, news articles, and stock quotes. A desktop personal computer with a traditional desktop browsing (scan, search) program communicates with the Internet, is directed to information retrieval (typically allocated to so-called pages), and lists Shows improved search results, allows selection from lists, displays text, images, multimedia representations (sounds, animations, videos) and links to other pages. A standard page is created by mixing the displayed information, especially the items displayed and the various symbols of the markup language that indicate the relative size, color and relative position of the information. The markup language can be read by a browsing program, for example, by word-wrapping large amounts of text in appropriate margins, so that the content presented can better utilize a particular display on a personal computer.

[0003] Recently, portable devices that can communicate via the Internet or operate like a telephone, have a relatively small display area, and have a relatively small keyboard have been introduced. Although such devices can follow a browsing process similar to the desktop-type browsing process described above, information received from the Internet is not designed to be presented on a small display or sound, There is no convenient way to achieve the presentation of this information without utilizing the present invention. For example, if the information is presented entirely in speech, the words may simply be recited from left to right as they would appear on the display of a desktop personal computer, which can be redundant and confusing to the listener. Let me do it. Further, there is no convenient way for the user to select, enter the requested data in some form, and reference hypertext to other pages (referred to herein as links).
It is desirable that these functions can be performed from a general telephone, that is, a device having only voice input and output capabilities.

[0004] Text and screen readers are known, and such readers have been used to present textual information audibly via synthesized speech. However, since the content received from the Internet is received as a mixture of text and markup language symbols, it is difficult for the average user who is not accustomed to markup language symbols to understand the received content in speech. Let's be. Since markup symbols define the visual organization of the text on the monitor screen, simply excluding the markup symbols destroys the relationships between the elements of the information, thus confusing the presentation of audio information. Would.

[0005]

While the Internet is inherently designed for visual-based access and navigation (eg, searching and following links), it is spoken or primarily spoken with a visual form. There is a growing interest in accessing information that is organized for presentation. For example, users can be visually impaired,
And individuals who would like to access the Internet via a common telephone depending on their preferences, timing, location, or proximity to the device. So for audio,
Alternatively, it would be desirable for individuals accessing the Internet content via a device primarily compatible with audio with a visual format to make the experience more understandable and controllable.

[0006]

SUMMARY OF THE INVENTION A system for audio browsing, or transcoding,
Information can be accessed and navigated through an audio user interface or a limited display user interface. The voice user interface allows voice or telephone keypad input to interrupt the presentation of information, for example, to directly follow a hypertext link. When operating over the Internet, a user of one or more access devices, including a mobile phone, telephone, portable computer, or workstation, can request a web page by entering an address. The user who operates the editor can create guidance in advance. Guidance can indicate summary, selection, annotation, and restatement of requested content. The guidance includes a description of the structural summary of the content, or a statement in a markup language that includes a hidden Markov model. In order to be able to frequently change the composition of content on web pages served by the Internet,
Portions of the content are identified for utilizing the specific guidance by aligning the structural summary description of the current content with the structural summary description stored with the guidance. Guidance appropriate for specific content is the content address (e.g., URL),
It can be located based on an associated address, or being within a regular expression. The guidance creates a derivative content that is easy to understand when presented by voice or easy to understand when presented on a display with a limited area or resolution. Derivative content may include a set of web pages organized hierarchically with connecting links. The complexity and importance of the area of the requested web page is assessed by revealing the hierarchy and organization of the set of web pages presented. Next, each web page is represented in a voice markup language and presented to the user via a voice engine. The information from the requested web page may be presented on the same access device or on another access device.

The presentation by the first user interface may include a link for transferring control to the second user interface. For example, a user of a limited display device, including a telephone (eg, a so-called web phone), may terminate the current session of the limited display user interface, place a call, and terminate the session with the audio user interface. The link can be followed on a limited display to establish.

[0008]

DETAILED DESCRIPTION OF THE INVENTION The present invention specifically provides a system for accessing and presenting information. The information is accessed via a computer network (eg, a local area network, intranet, or the Internet). The information can be accessed using a regular telephone or computer workstation. In addition to presenting information audibly, the user can respond to words or phrases conveyed by one or more voices (such as a female voice or a male voice), and thereby respond (eg, to navigate the Internet). You can request different information. To request information, the user may repeat the word, activate a button on the telephone, or say a command. The response thereto specifies the link to follow, or specifies information to provide to a processing process or server on the network.

A system in accordance with various aspects of the present invention includes a processing process for transcoding, audio browsing, and text-to-speech software for conveying information in speech for a user to hear. Transcoding is generally a markup language (eg, hypertext markup language (HTML), extensible markup language (XML), and wireless markup language (WML), and speech extensible markup language (VoiceXML). Are the processing processes that operate on the adaptively represented information of these systems for voice and wireless systems, such as, for example, the results of transcoding are spoken and input to a text-to-speech processing process. Voice browsing may be a response to any of a variety of inputs from the user, including the words spoken by the user and keystrokes on a keyboard or keypad for telephone dialing. Including following links to pages That.

Operating a system according to various aspects of the present invention provides the user with: (A) access to information such as news headlines, financial information, weather information in a personal and substantially user controlled manner; (b) substantially "voice" of such information. Only "version. Such a system allows a user to search for customized news and other information. In addition, information requested by the user (eg, a web page) that has not been edited by the user can be accessed. For example, information received from the Internet (e.g., a web page in HTML) is transcoded by a guidance application and the selected information is transmitted to VoiceX so that it can be communicated to a user via a speaker or telephone.
It can be provided in ML.

Services operable in conjunction with the Internet and provided in accordance with various aspects of the present invention include, among other things, personal pages for each customer, and conventional viewable HTML web pages. A transcoding service that transcodes to a set of pages where you can listen to Each customer has e.g.
-Specify personal pages including emails, calendars, news reports, sports reports, weather reports, traffic reports, and movie listings.

Any information in hypertext or markup languages (eg, Internet web pages) can be transcoded in accordance with various aspects of the present invention. Transcoding also includes applying the instructions provided in a file, referred to herein as guidance. Guidance can be created by the user for each web page of interest and easily tailored to the preference when listening. Guidance includes volume, speed,
Specifying characteristics such as pitch and direction, suppressing output for specific elements, character-by-character, text inserted before and after elements of audio fonts (male / female, adult / child, etc.) content,
By specifying features such as sound effects before, during or after content elements (spell out certain text) and the addition of music or pre-recorded sound, the web page The aspect to be displayed (eg, visual and / or audio) can be specified.

In accordance with various aspects of the invention, a communication system may include a plurality of computers (eg, servers, Workstations and portable devices including telephones and mobile computers). This system is especially useful for graphic users
An interface (GUI), a voice user interface, and a limited display user interface can be provided. For example, the system 100 of FIGS.
Includes an access device 110, a gateway device 120, a network 130, and a network server 140. The access device 110 is connected to the gateway 120 at any time via any conventional physical and logical protocols. The gateway 120 and the network server 140 may be timely (or by a dedicated link) including physical and logical protocols that may be different from the physical and logical protocols described above in connection with the access device 110. Connected to the network 130 via any type of network technology. These protocols allow information to be transferred simultaneously and efficiently between either one or more gateways and servers. Each functional block in FIG. 1 operates individually (e.g., like a cell phone operated by an individual user) and serially (e.g., multiple transcoder gateways working together from a common input and output queue). Any number of similar operating at different geographical locations (such as wireless gateways for home site servers at different protocols or different affiliated headquarters), or different special purposes Represents a functional block.

An access device includes any device that facilitates accessing information stored on a server connected to a network. Access device 110 comprises, for example, an unrelated set of various devices that are operated separately by many users. Exemplary devices include a workstation 111 (eg, a desktop computer), a wireless device 112 (eg, a laptop, palmtop, or network device) phone 113, and a mobile phone 114 (eg, any wireless phone or pager). Contains. The links shown between the functional blocks are, for example, wireless,
Includes any communication capabilities including wired networks, including infrared and cable TV, local area networks, public telephone networks, and proprietary leased lines.

The workstation 111 has a link 170
And 179 allow connection to the network 130 via an Internet Service Provider (ISP) gateway 124. Wireless device 112 establishes wireless link 17 to wireless gateway 121.
1, a link 172 to the ISP gateway 124, and a connection to the network 130 via a link 179. Alternatively, the wireless device 112 has a wireless link 171 to the wireless gateway 121, a link 174 to the transcoder proxy server 125,
And a link 178 to the network 130. Phone 113 can connect to network 130 via a wired link 173 to voice browser server 123, a link 177 to transcoder proxy server 125, and a link 178. Cellular phone 114 is wireless link 175, cellular gateway 12
2. Link 176, link 177 to voice browser server 123, transcoder proxy server 12
5 and link 178 to network 130.

Communication between the functional blocks shown for system 100 may utilize any of the conventional protocols. Each of the logical and physical protocols for each of the aforementioned links is implemented with some of the protocol functions at the end of the link. Table 1 below shows the appropriate protocol functions.
It is described in. The gateway 120 may cooperate with the access device 110 to support connection service or connectionless service.

[0017]

[Table 1]

[Table 2]

[Table 3]

[Table 4]

The gateways and servers typically include computers with storage capabilities and / or special purpose hardware and software for efficient data access and network communication. For example, the functional blocks of each of the gateway 120 and server 140 may be a disk storage subsystem that conforms to industry standards and conventions for use on the Internet.
Communication interface (eg, modem, wireless transceiver)
And a plurality of computers having a network interface (for example, a T carrier). In another embodiment, the workstation 111 comprises devices similar to the gateway and server devices, and vice versa. In other words, gateways, servers, and workstations are distinguishable by the functions performed, which are primarily based on the software described above, which runs on an otherwise general-purpose computer.

Each computer in system 100 (including those embedded in wireless devices, telephones, and cell phones) has functions implemented in software (including firmware, for example), collectively referred to herein as processes. Can be included. A process can include an operating system and a set of application programs suitable for the function being performed. The application program may include processes specific to the hardware interface (eg, drivers) as well as several processes or processes suitable for simultaneous use by threads. The process may be integrated with the operating system or may be loaded and saved separately from the operating system. The process including the operating system is implemented in firmware and, depending on the function, in a circuit.

The sequence of messages provided by a process operating in a system according to various aspects of the present invention may comprise a voice user interface. In particular, the voice user interface allows a user to access information stored on the network (eg, Internet navigation by following hypertext links). For example, the message sequence 200 of FIG. 2 cooperates with the process to receive a telephone call from a user of the audio device and serve an Internet web page via a link identified by the user. To clarify the description of the voice user interface, the Internet and its conventional protocols perform the functions of the network 130.

The audio device 202 has an audio input / output (I /
O) Represents any device that has the capability and ability to participate in voice telephone calls. The calling process 204 performed by the voice device 202 includes the voice browser server 1
Perform all conventional steps for initiating a telephone call to 23. The answer process 208 is performed by the voice browser server 123 to perform all the conventional steps, and after the call to receive the call has been established, the call process 204 causes the answer process 208 to perform the conventional login scenario. Identification sufficient to complete (step 244) (eg, via a user initiated DTMF signal or automatically by calling process 204)
Is performed. The login process 212 performed by the ISP server 124 receives this and responds to the login scenario with an IS for the user of the audio device 202.
A session with the P server 124 is established. Any of the conventional processes can be used for the calling process 204, the answer process 208, and the login process 212. In another embodiment, the session is implied by a system operation (eg, a dedicated link) and steps 242 and 24
4 is omitted.

Processes 216-226 may be performed by five servers at separate physical locations as shown. In another embodiment, the process 216
-222 may be performed by a single server at one location. Preferably, it is a separate server to increase processing power for each of the processes 216-222. Processes 212 and 222-226 include conventional software, such as Apache software
It can include the Apache Web, which is commercially available from the Foundation (Apache Software Foundation). Processes 204 and 206 may be performed by conventional analog telephone circuitry.

At some point after the session has been established (eg, in response to messages between processes initiated by response process 208), voice browser server 12
Audio Browsing Process 21 Performed by 3
6 requests information expected to be provided by server 140 over network 130 (step 24)
6). The first request may correspond to a "home" page. The request can be made by addressing the message to a transcoding process 218 that identifies the requested page. In the Internet embodiment, the request includes a uniform resource locator (URL) with an argument of the type described in RFC 1738 (a comment request available from the W3C) (URL1 at step 246). URL1 contains the address of the transcoder proxy server 125, the ISP server1
24, the address of the home site server 141, the page requested by the home site server 141, the address of the voice browser server 123, and an indicator identifying the type of the process 216 as an agent.

Transcoder proxy server 125
The transcoder process 218, executed by, generally restates the information from URL1 to prepare URL2 (step 248). URL2 identifies the address of the transcoder proxy server 125, the address of the ISP server 124, the address of the home site server 141, and the pages that are sufficient to obtain the desired home page requested by the home site server 141. A label may be included.

The routing process 222 performed by the ISP server 124 performs conventional routing of requests. For example, the routing process (eg, HTT
URL (according to P and TCP / IP protocol)
2 and sends a request (substantially the same as URL2) to the serve process 224 (step 250).

The serve process 250, performed by the home site server 141, performs conventional search, access, and response functions to satisfy incoming requests.
In the case of the Internet embodiment, serve process 2
24 provides a web page (PAGE1) containing information in a markup language with tags (step 25)
2). PAGE 1 is typically suitable for presentation on a workstation with a relatively large screen, color monitor.

The routing process 222 is a process 224
Transcoding process 2 when PAGE1 is received from
18 (step 254) routing process 222
Does not normally revise the information sent.

Upon receiving PAGE 1, transcoding process 218 creates a replacement page (PAGE 2) and transfers it to audio browsing process 216 (step 258). The operation required to create PAGE2 is generally called transcoding. P
AGE2 summarizes information from PAGE1, and PAGE2
Information, annotations, and PAGE 1
Restatement of the information.
PAGE2 may be provided in any markup language. Preferably, the transcoding process 218 utilizes the identifier of the audio browsing process 216 to determine an appropriate markup language. For example, if the voice browser server 123 has one or more browsing
If multiple browsing processes are running, and one browsing process accepts HTML while another browsing process accepts XML, the transcoding process 218 will use the identification indicator received at URL1 as described above. PAGE2 is created in a markup language corresponding to.

The transcoding process 218 summarizes the information provided in step 252 if the amount of information exceeds limits in terms of size and / or complexity. The resulting summary may be hierarchical. For example, PAGE1 has 4 news, national and local weather,
And PAGE if average stock market height was included
You can create a set of pages with 2 added. PAG
E2 may include only a headline for each news, followed by a link to the full news, a link to local weather, and a link to the average stock market height. Other pages in this set include, for example, PAG
E2.1 (news), PAGE2.5 (national weather report), PAGE2.7 (list of average heights), PAG
E2.7.1 (Dow value, mean end value, mean high value, mean low value) and PAGE 2.7.2 (NASDAQ value,
(Average end value, average high value, average low value).

The information can be selected from the information received in step 252. In other words, the information is PAGE2 (or a lower level page of the set as described above)
You can also prevent it from appearing. Such information is stored in device 20
2 may not be suitable for the user interface used in (eg, not suitable for sound or too large for small display areas). For example, the description of the PAGE1 bitmap element may be omitted. Or, to avoid confusing users with unnecessary redundancy, PAGE
A description of a link that appears in several places in one may not be required.

Annotations can be added before and / or after the element of information received in step 252. For example, if PAGE1 contains a daily Dow Jones average declared +139.4, then PAGE2
Can include information describing the numerical value of 139.4 as "139.4 points higher" as follows.
The word "high" may be replaced with "+" and rephrased as "139.4 points higher." Voice device 2 by annotation
02 can recognize the information more accurately.

The reproduction of the information can include conventional transcoding operations as well as the operations necessary for the user of the device 202 to more accurately recognize the information. For example,
Animation or marquee on PAGE1
If text is included, PAGE2
Regardless of the original animation or scrolling presentation at 1, the text can be created to be completely recreated. Additionally, PAGE1 may be an inline script (eg, style information or a reference to a cascading style sheet), or program code information (eg, a JAVA applet,
Or a common gateway interface (CGI))
, The transcoding process 218
Can create part of PAGE2 to represent a suitable alternative function compatible with the audio browsing process 216.

The transcoding process 218 is instructed to perform, in particular, summarize, select, annotate, and reproduce according to the guidance created, stored, retrieved, and referenced in accordance with various aspects of the present invention. . For example, the guidance represented as MASK database 230 in FIG. 2 differs from a conventional style sheet in several ways. Conventional style sheets (eg, written in CSS for HTML or XSL for XML) must be specified by references that appear in information (eg, part of PAGE1). In contrast, in accordance with various aspects of the present invention, PAGE 1 need not include a reference to the guidance of database 230.
The browsing process receives the content for presentation and translates the markup language tags of the received content with reference to a conventional style sheet. In contrast, the guidance in database 230 instructs the transcoder process how to create the derivative content to be input to the browsing process. The transcoding process receives content for transcoding in a first markup language and how to produce output to a browsing process in the same or a different markup language than the first markup language. Refer to the guidance in the second markup language to indicate what to do. As a result, according to various aspects of the present invention, a second markup language is not possible with respect to using a conventional style sheet.
Or, a tag indicating a meaningless function is included.

The transcoding process 218 is an ISP
Upon receiving an index (eg, URL1) identifying desired information obtained via server 124, or
When an index (for example, the title of PAGE1) for specifying the information received from the server 124 is received, an appropriate guidance is searched from the database 230. (Step 2
56 to 258). As described above, summarization, selection, annotation, and reproduction are then performed based on the information of PAGE1,
PAGE2 is created. PAGE2 can be adapted to a voice markup language (eg, VoiceXML or proprietary voice markup language). Next, the transcoding process 218 executes PAGE2
To the audio browsing process 216 (step 260).

The audio browsing process 216 can include conventional text-to-speech (TTS) capabilities with input queues and analog audio output capabilities. For example, the audio browsing process 216 may be Micr
Micro Marketed by Microsoft Corp
soft SAPI. The audio browsing process 216 may include incoming information (eg, P) sent from the transcode process 218 in a markup language.
AGE2), replace and add corresponding information based on conventional audio style sheets and other configuration settings, load the TTS queue, and send the analog signal (RE) to the I / O process 266 of the audio device 202.
CITAL1) is provided (step 262).

The audio browsing process 216 may include a DTMF signal from the audio device 202 and / or a device 2
02 (Step 2)
64). Speech browsing process 216 may include conventional speech recognition (SR) capabilities with analog speech input. Audio from the user causes the I / O process 206 to send the analog audio signal SPEECH1 to the audio browsing process 216. For example, the I / O process 206
Sets the speaker of the audio device 202 to "1
Press or say 1 "voice message, and then the user responds by saying" 1 ", the I / O process 206 causes the analog audio signal (SPEECH1) corresponding to" 1 " The voice browsing process 216 responds accordingly to determine the URL to send the local weather information (step 266),
And the URL (along with the information described above with respect to step 246) (URL3) transcoding process 218.
Send to

The transcoding process 218 determines an appropriate URL for local weather, similar to the operation performed as described above in connection with step 248, and
(Eg, URL4) to routing process 222 (step 268). The routing process 222 is similar to that described above in connection with step 250,
L (eg, URL4) to any of the serve processes 226 of the site server 142 (step 27).
0). The server 142 responds to the information in the markup language (eg, PAGE 3) (step 272) and is sent to the transcode process 218 via the routing process 222 without substantial revision (step 274).
The transcoding process 218 creates information in a markup language (eg, PAGE4) and sends it to the audio browsing process 216 (steps 276-28), as described above with respect to steps 256-260.
0). The audio browsing process 216 includes step 2
62, the analog audio signal (R
ECITAL2) and sends it to the I / O process 206 (step 282), which provides the user with information (eg, summaries, selections, annotations, and reproduced national weather information) corresponding to the requested page to the server. 142.

According to various aspects of the present invention, a method for providing a user interface provides information adapted for audio presentation and / or adapted for restricted display. Such user interfaces receive commands verbally or as keystrokes to navigate to other information for presentation. Thus, the computer starts with a request for content from another node of the computer network from FIGS.
Can be performed (step 302). Content means some form of information. The content is provided in a message (eg, a page, streaming audio, or streaming video) or an object (eg, a file, or downloaded executable program, applet, or script). The content may be a home page on the Internet as described above. The request may originate at step 246 based on any conventional protocol, including by providing a URL as described above.

Derivative content is created based on at least a portion of the so-called primary content requested and received in a previous step (step 304). Derived content may be stand-alone (eg, a single message or file) or flowing (a series or set of messages or files). The derivative content is in the form of a page (for example, P
AGE2). Is the guidance for creating derivative content integral to the primary content (for example, in a conditional inline style) or is the primary content a reference (for example, an external style sheet)? , Or separate from the primary content. In one embodiment, the primary content is not guidance-based. For example, as described above, stand-alone derivative content (eg, PAGE2 or PAGE4) is separate from primary content (eg, PAGE1 or PAGE3) and is based on guidance (eg, database 230) that does not use this as a reference. May be created based on the

The derivative content can be presented while allowing the interruption of the command (step 306). Commands can be directed to the user to whom the presentation is directed, to monitor the presentation for someone else, or to an automated process (for example, based on external criteria such as number of visitors, number of attendees, missing attendees, or time). Automatically order the devices used for advertising, alerts, entertainment, or teaching). The interrupted command is sent to the keypad (eg, 113 or 1) as described above.
14), a keyboard (eg 111 or 112),
Or voice (eg, 111, 112, 113 or 1
14) can be started. In one embodiment, one or more access devices are used simultaneously by the same user. For example, the monitor of the workstation 111 can receive presentations that can be used simultaneously by the same user of the device 112, 113, or 114. The command may be of the type described in Table 2.

[0041]

[Table 5]

[Table 6]

If the presentation is complete (audio or visual) and the user command does not interrupt the presentation, the method may be used for an appropriate amount of time to give the user the opportunity to issue any appropriate command as described above. The next step can be delayed (step 308). When the delay time elapses or the command is entered, the command (or default command (default command)) in the context of the current presentation (eg, "return" is to the current presentation) (Step 31)
0). When the command is executed, the method is repeated starting at step 302.

The creation of the derivative content is achieved according to various aspects of the present invention, based on a method comprising one or more of the following steps. That is, a step of acquiring the content for presentation based on the guidance, a step of finding the position of the node based on the guidance, and a step of reducing the complexity of the presented content. For example, step 304 of method 300 can be accomplished by the coordination of several processes illustrated by the data flow diagram of FIG. 4 to create derivative content 440. The process 304 receives as input the primary content 400 (eg, one or more files, buffers, or messages as described above) and the address of the primary content (PC), eg, a URL (PC URL).
Receive. Process 304 is guidance process 40
2, finding the model skeleton process 404; analyzing the primary content 406; reducing complexity 408; creating the primary content skeleton 410; aligning the skeleton process 412; It includes a process 422 for obtaining a reference, a process 424 for finding a node, a process 430 for obtaining a node by a skeleton position, a process 428 for obtaining a node by a node name, a process 432 for obtaining a node by matching contents, and an annotation process 436. . Each process 402-436 may be executed by the transcoder proxy server 125 from time to time (eg, serially or in parallel) at any time when data is appropriately available for that process.

The process 402 of finding guidance is P
Upon receiving the C URL, obtain appropriate guidance relating to at least a portion of the PC URL, or obtain default guidance if the appropriate guidance is not associated with the PC URL. Preferably, the guidance is obtained in a markup language, for example, a MASK markup language. The guidance found in this way is sent to the next reference acquisition process 422 and annotation process 436.

The find model skeleton process 404 receives the PC URL, obtains an appropriate description of the model content associated with at least a portion of the PC URL, and associates this description with the alignment process 412, and the skeleton location to identify nodes. Send to acquisition process 430. The description is preferably stored in the form of a tagged skeleton in the MASK markup language. The skeleton may be included with guidance for convenience of access. In such embodiments, the model skeleton discovery process 404 is omitted and the guidance discovery process 402 supplies the skeleton as needed for other processes.

When a description of the model content is found, it can be provided as record 415. The description of the model content includes an association between the descriptive identifier and a node identifier for each node of the model content. For example, the record 415 provided by the model skeleton discovery process 404 includes, for each model node, a single printable character as a descriptive identifier (MODEL LETTER) and a node identifier (MODEL).
NODE) can be included.

The method used to find guidance need not be the same as the method used to find the appropriate description of the model content. In one embodiment, the guidance includes a reference to the appropriate description of the model content, and much more between the record (or file) containing the guidance and the record (or file) containing the description of the model content. Promote the relationship. In addition to this, access to either the guidance or the description of the model content may be provided, for example, by the user for the type of device (or device having a particular identifier) utilized in step 302 for which the derivative content is created. May be based on information from the user account, including the user's preference.

A primary content analysis process 406 reads the primary content 400 and a PC skeleton process 410
And get node 428 by node name
Create a record 418 that is used by the process 432 used by the process 432, a node discovery process 424, and a complexity reduction process 408 that are used by the process 432 to obtain nodes by content matching. Records 418, 419, and 420 can be organized in any conventional manner, for example, as included in a type of database known as a document purpose model (DOM). The DOM can be stored as a conventional database (eg, a relational database or a star database). Information in the form of a DOM can be stored in memory in any suitable storage format, including arrays, linked lists, graphs, trees, data structures, combinations thereof, and the like. Typically, DO
M represents a data structure representing a tree, or nodes, branches,
And a graph with leaves. A leaf (eg, element) can be represented as a data structure consisting of any object (eg, text, table, bitmap, applet, link, etc.) that is expected to be included in the content. Nodes and branches can be represented as data structures consisting of lists (eg, pointers to elements, or other lists). One node is typically called the root for accessing any leaf. Access to specific content (eg, leaves) begins at the root (“0”) and is followed by tracing pointers (crossing branches between nodes) for each part of the node name (eg, fourth at node “0”). , Followed by the third pointer in the list of nodes "0-4"), the node name (e.g., "0-
4-3 ") can be achieved with reference as a reference.

The analysis process 406 determines, for each node of the primary content, a node name (eg, PC NOD).
E. Determine the relevance between "0-4-5" and a part (for example, a table) of the primary content. The analysis may proceed based on the nesting level of the tag in the markup language of the primary content 400. These associations can be stored as record 418. Each record 418 may include a copy of the associated content from primary content 400. The analysis process 406 determines, for each node of the primary content containing text, the association between the node name (eg, “0-4-3”) and any words (or other significant symbols) used in the text of that node. Is determined. These relevances can be stored as record 419. Record 419 may be stored in a form known as a hash table. Each record in the hash table collectively has a hash key for a particular word (or significant symbol) of the primary content,
And a list of node names for each node that has content consisting of specific words (or significant symbols). The analysis process 406 determines, for each node of the primary content, a set of numerical descriptions (herein referred to as node statistics) and a node name (eg, “0-4-3”) and a set or set components. Form a relationship. These relevances can be stored as records 420. Node statistics include quantitative <no
As described below in connection with destat>, for example, any quantitative property may be included.

If the guidance process 402 does not find suitable or default guidance,
Creation of derivative content 440 is recorded in records 418 and 42
Go to base 9 Record 418 (eg, first from root)
For each node (depth of order), the complexity reduction process creates the derivative content as one or more nodes of the derivative content. Derivative content 440 is preferably provided in a markup language. The copy of the primary content associated with the node in the record 418 can be identified as derivative content, or copied as a derivative content as a file, buffer, or message. If the content at the primary content node is too complex, additional or replacement content may be created for one or more nodes of derivative content.

If the guidance discovery process 402 finds a suitable guidance location, the preparation of the derived content 440 based on divergence may still produce results that are too complex to present. If the record 418 matches such guidance (eg, if so indicated by the annotation process 436), the creation of the derivative content 440 proceeds. The complexity reduction process 408 can read the record 418 and create derivative content 440 as described above.

In each case, the complexity reduction process 40
8 determines the complexity of the content at the node with reference to the quantitative properties of the node. For example, record 4
The quantitative properties described above in connection with 20 can be read by the complexity reduction process 408. The complexity reduction process 408 can compare such a quantitative property to a predetermined threshold (eg, a limit). The derivative content provided by the complexity reduction process 440 primarily includes content that has been tested or created and determined not to exceed appropriate limits.

In accordance with various aspects of the present invention, comparisons are made based on primary content and model content, particularly to facilitate identification of derivative content. This comparison may be made between the description of the primary content and the description of the model content. Preferably, the comparison is made between the primary content skeleton and the model content skeleton. The primary content skeleton creation process 410 creates a description of the primary content in a format suitable for comparison. For example, the model skeleton discovery process 404 provides the record 415 as a skeleton as described above, and the PC skeleton creation process 4
10 supplies a record 416 consisting of the corresponding information determined from the record 418. The description of the model content includes the association between the descriptive identifier and the node identifier for each node of the model content. The record 416 provided by the PC skeleton creation process 410 includes, for each primary content node, a descriptive identifier (PC LE
A single printable character as the TTER) and a printable hyphenated digit string as the node identifier (PC NODE).

The guidance is created with reference to the model content node. Correlation between records 415 and 416 is performed to obtain the corresponding node of the primary content. This correlation is performed on the entire record 416 using the entire record 415. For example, in embodiments having a single character for each descriptive identifier, the descriptive identifier for the model may be aligned with the descriptive identifier of the primary content as a correlation between the character strings. The skeleton registration process 412 generally associates a PC node with each model node. (A) PC skeleton with nodes not found in the model skeleton, (b) model skeleton with nodes not found in the PC skeleton, (c)
Consider inconsistencies such as model nodes not found in the same order as provided in the PC skeleton, and (d) PC skeletons with nodes (and all subordinate nodes) recognized as corresponding to the model skeleton. Conventional processes can be used. The alignment process 412 is based on the model
Gives an association between the node identifier and the PC node identifier. This association is provided as records 417, where each record is a model node identifier (eg, "0-2-6").
And the primary content node identifier (eg, "0-
4-3 ".

The relationships described above in connection with records 415-420 may include any one or more conventional methods (combinations, combinations) including tables, arrays, or linked lists in memory.
Or separately) or as a file or database on the secondary storage device.

The guidance sets the primary content node to 1
Derived content can be specified by referring to it once or multiple times. If guidance is provided without a description of the model content (eg, without a skeleton of the model content), the reference criteria should include words (or significant symbols) suitable to identify the primary content node. You may. If guidance is provided with a description of the model content, the reference criteria are the node identifier (eg, “0-4-3”), and the location for the description of the model content (eg, skeleton position (SP)).
Can be included. In other words, for each node of derivative content to be created, the guidance may include one or more references for obtaining the content of the node of the primary content to be included as derivative content. For each node of the derivative content, the next reference retrieval process 422 provides a set of references to the node discovery process 424. Preferably, the set includes several references, so that (a) if the use of one reference does not identify a node of primary content, an alternative reference can be successfully used; And (b) if the reference criteria identifies more than one node of the primary content, or if the set of reference criteria identifies more than one node of the primary content, make a determination to analyze ambiguity. be able to. If the node discovery process 424 achieves the designation of the next node in the derivative content, the node discovery process 424 works in any conventional manner with the next reference acquisition process 422 to process the entire set. Get each set of reference criteria until complete.

For each set of reference criteria received from the next reference criteria acquisition process 422, a node discovery process 4
24 identifies the node of the primary content to be included in the derived content. In one embodiment, such identification may be performed by associating a flag with one of the records 418. In another embodiment, the organization (eg, order or tree structure) of the derived content may be different from the organization of the primary content, as the flag may specify the number of nodes of the derived content.
The one or more references provided by the next reference reference acquisition process 422 include the node discovery process 42
4 can be provided in a conditional configuration to facilitate if-then-else or switch (eg, case) operation. For example, the conditional construction allows the node discovery process 424 to utilize a first set of references if the condition is true;
If the condition is false, a second reference (eg, omitted) can be used. Conditional logic can refer to node statistics 420 as described above. The node discovery process 424 obtains a node identifier for each reference,
Analyze ambiguity between node identifiers to determine if a node was found, and if so, which nodes to include in the derived content (if some nodes are applicable) Is specified. Embodiments for analyzing ambiguity are described below after processes 428-432.

The find nodes process 424 can receive references to the nodes of the model from the next get references 422 process. The reference criterion may be any conventional type of node identifier, including those of the form ("0-3-2"). Process 42 for acquiring the name of each node
8 receives such a node identifier and obtains the associated PC node by querying (or referring to) record 418. In other words, if the current primary content has a structure that includes nodes named as being in the model content, then the primary content node is retrieved by the process 428 to obtain a node-by-node name. Is supplied with the node of the primary content. If the node by that identifier is not present in the primary content, the resulting display is returned instead. If the set of references provided by the next reference reference acquisition process 422 includes one or more such node identifiers, the retrieve node-by-node name process 428 will return the corresponding query (or reference). And return each result.

The find node process 424 can receive references to the next part of the description of the model content from the next reference reference acquisition process 422. For example, if the description of the model content is a skeleton (eg, a character string) as described above, references to parts of the skeleton (eg, substrings) should include a character start position and a character end position. Can be. The process 430 of obtaining nodes by skeleton location attempts to identify corresponding parts of the primary content description. For example, record 41
A query or reference to 5 can provide one model node identifier associated with the character start position and another model node identifier associated with the character end position. A query or reference to record 416 can provide a PC node identifier for each model node identifier. Thereafter, depending on the skeleton position, the process 430 for obtaining nodes returns one or more PC node identifiers as a result. P
If the C node identifier constitutes a tree, the root of the tree can be returned. If none of the PC node identifiers correspond to a skeleton position, the resulting display is returned instead. If the set of references provided by the next reference reference acquisition process 422 includes one or more such skeleton locations, the 430 obtain nodes by skeleton locations process 430 may be performed by a corresponding query (or reference). ) And return each result.

The find node process 424 can receive reference criteria as words, symbols, or phrases (eg, character strings) from the next reference reference acquisition process 422. For example, content can be identified by a table title, such as in a “market snapshot”. The get nodes 432 by content match process 432 attempts to identify corresponding PC nodes that match words, symbols, or phrases. The process 432 of obtaining nodes by content matching forms a hash equivalent of all or some of the reference criteria and queries (or references) on record 419 to obtain PC node identifiers.
I do. If none of the PC node identifiers contain the referenced content, the result is returned instead. If the set of references obtained by the next reference reference acquisition process 422 includes more than one such reference to the content, a process 432 of obtaining nodes by matching the content corresponds. Perform a query (or reference) and return the respective result.

As dictated by the guidance, the node discovery process 424 can analyze the relative references for any returned PC node identifiers. For example, the guidance may indicate that the parent (of the reference-based node) is intended (e.g., by syntax such as "<-" below). Node discovery process 424 is performed as described above for process 428-4.
From 32, zero or more PC node identifiers can be received. If no PC node identifier is returned, the find node process 424 does not create the node information to add to the annotation process 436 and proceeds to obtain the next reference from the next reference process 422.
If one PC node identifier is returned, the node discovery process 424 can submit the PC node identifier to the annotation process 436. One PC node identifier corresponding to the model node (M) is found and more than one P
If the C-node identifier is returned from processes 430 and 432, the ambiguity can be analyzed in the following manner. That is, 1. It takes into account that each reference creates a corresponding list of PC node identifiers. 2. A list of PC node identifier candidates (C _{1... K} ) is formed, where each candidate (eg, C ₁ ) appears on all lists. 3. A score is calculated for each candidate, and a PC node identifier having the smallest score is selected. Each score can be calculated by the following formula.

[0062]

(Equation 2) However, C _n is a candidate score is S _n. Each C
Has a P level in its node name (eg, "0
The node name "-0-3-5" has four dimensions, one for each of the four levels.) M is a model node with a level of Q in the node name. , "Min ()" return the minimum value of its argument.

The score represents the “distance” from the model node to the candidate node. The distance is calculated using the square root of the sum of the squares of the dimensional differences. If the levels of the candidate node identifier and the model node identifier are different, each level that does not appear in both the candidate and model node
An arbitrary dimension difference of 0 is assigned. In another embodiment,
Any dimensional difference in the range of 2 to 20 is used.

When one PC node identifier is determined by the node discovery process 424 ("node discovery"),
The notification of the node discovery is communicated to the annotation process 436.
Annotation process 436 can send the revised content to complexity reduction process 408 for use as derivative content. The revisions made by the annotation process 436 may include summarization, selection, annotation, and / or reproduction of primary content related to node discovery. Revision can be accomplished based on any conventional editing process appropriate for the content being revised. For example, additional text or audio can be added to the text content.

The guidance can dictate how complexity and importance are determined and, as a result, how derivative content is prepared. In other words, the guidance can specify the specific measures and limits used to determine complexity, and how the average user responds to content, that is, what information is most important, Or derived content can be developed based on knowledge such as what to consider least important, clear understanding, and spelling rather than audio to facilitate navigation.

In accordance with various aspects of the present invention, the guidance discovery process includes any process that accesses the guidance utilized by the transcoding process. For example, the guidance discovery process 40
2 can include the method of FIG. 5A to limit whether to apply the default guidance. In such a manner, default guidance is used when no other guidance is available as relevant to the primary content. Such default guidance includes threshold condition or limit metrics used to analyze the content (eg, measures and limits for content complexity, measures and limits for expected screen size, content Characteristics that are tested to recognize the organization). Default guidance can be accessed based on the results of such an analysis.

The identification of appropriate guidance can be achieved according to various aspects of the present invention, based on a method comprising the following steps. That is, access to guidance based on primary content addresses, addresses associated with primary content addresses, and addresses within the range of regular expressions, and identify such guidance as appropriate for the primary content area. , One or more steps that correlate at least a portion of the primary content with content relevant to the guidance.

In one embodiment, the guidance is the primary content address (eg, the primary content URL) 0
Is related to Using addresses as criteria,
A database can be queried that provides one or more records (eg, files) with guidance indicators.

In another embodiment, the query is made by replacing the regular expression with part of the address of the primary content. As an example intended for use with the Internet, content is addressed by a URL containing a date. By including a date, an archive of the content can be stored and accessed by navigation. A web site that distributes news can provide new content on a daily basis by addressing each page with a date code in the YYMMDD format for each year, month, and date. Guidance applicable to content regardless of the date code may be omitted from databases with queries that omit the date code or reproduce the date code as a regular expression to avoid preparing guidance that is also addressed by the date code. Can be obtained. In other words, when a user requests content addressed by a date code, an address suitable for the query can be derived from the requested address by replacing the regular expression with the date code. According to various aspects of the invention, any portion of the primary content can be replaced with a regular expression. Several regular expressions can be substituted to form a single address suitable for the query.
Examples of regular expressions are given in Table 3. Many regular expressions with varying syntax are known. The so-called wildcard characters constitute a simple kind of regular expression (eg, the file name syntax " ^* " and "?" Supported by MSDOS available from Microsoft Corporation).

[0070]

[Table 7]

Content available through a computer network may change, for example, if different computers connected to the network are managed by independent authorities for different purposes. Changes in content that are understandable when viewed on a workstation monitor (eg, information is presented in a visually perceived manner with respect to the relative position and emphasis of the information) can be reduced when analyzed from the source markup language. Manifest as structural and / or aesthetic changes. Based on various aspects of the invention, predefined guidance appropriate for the content before the change follows the change to ensure that a consistent presentation is made on an audio device or a device with limited display Identified as suitable for similar content. As a result, the predefined guidance need not be revised to remain in a state suitable for developing content derived from the primary content that has changed to some extent.

In the preferred embodiment, the guidance is created from the model content, and the ITU-TX. 680 “Abstruct Syntax Notation
1), ASN. 1 "and RFCs 2251 and 22
52 is stored and accessed utilizing a directory access structure and method of the type known as the Lightweight Directory Access Protocol (LDAP) described at 52. Guidance accessed using LDAP is stored as an entry in the registry information tree. Each entry corresponds to a tree node. Attributes can be associated with nodes. Attributes include, for example: NAME-the identifier of the guidance entry at the node, for example a file name. 2. DESCRIPTION-description of guidance provided by the user. For example, a user who has prepared guidance using an editor of the type described below with reference to FIGS. 8-10 can describe the guidance in its own words for later reference. 3. OBJECT-the guidance itself. For example, M
It may be text specified in a markup language such as ASK markup language. 4. USER_ID—An identifier associated with the user (eg, during the registration process) and considered to correspond to the same user during any subsequent session. USE
By associating the R_ID with specific guidance, different users can enjoy different derivative content from the same primary content. Figure 8-1
Separate guidance can be defined using the editing process described below with reference to FIG. The value of USER_ID may correspond to any user group (for example, all users). 5. MODEL_ADDRESS-Guidance TEXT
The complete address of the model content used in the preparation of the file. For example, a complete URL. 6. APPLIES_TO_ADDRESS-The address to which the guidance OBJECT should be applied. For example, an address used in step 502, such as a primary address or an address consisting of one or more regular expressions. For example, the user who created the guidance OBJECT from the model content can also specify the same APPLIES_TO_ADDRESS as the URL of the model content or the URL consisting of one or more regular expressions to apply the guidance OBJECT more widely. 7. HAS_REGULLAR_EXPRESSION_
APPLIES_TO_ADDRESS is at least 1
FIG. 6 is a graph showing that one regular expression is included. 8. IS_FIRST—flag indicating that derivative content determined from guidance OBJECT should first be provided to the user interface. LD
A list of entries can be provided in response to an access request using the AP. Guidance OB of the entry
The derivative content determined by JECT will be provided with guidance O in another entry of the list before the derivative content.
Only one entry from such a list may have the IS_FIRST flag set to indicate that it is presented as determined from the BJECT.

The use of the directory information tree and the access protocol as described above facilitates the acquisition of guidance for preparing derivative content as a set of pages. Search for appropriate guidance performed by guidance discovery process 402 (step 502)
Is the appropriate APPLIES_TO_ADDRESS
Multiple entries (ie, nodes) with attributes can be returned. Such a search may look for entries based on one or more target attribute values. In one embodiment, the list of entries returned from an LDAP search can be further analyzed by a script (eg, regular expressions can be evaluated in the PERL programming language).
For example, the primary address may match the APPLIES_TO_ADDRESS attribute of each of several entries, or the primary address may have more than one A
It can be within the regular expression of the PPLIES_TO_ADDRESS attribute. If the search results identify one or more entries, the entries may be processed in any order (step 503). Returning to the calling process can occur after step 502.
Preferably, as described above, the entry has been identified as the first entry, for example, by the flag IS_FIRST. If the primary content is to be presented as a series of pages (eg, a set described below), the first entry may correspond to the first page of the set. If the primary content is to be presented as a hierarchical set (e.g., a summary described below), the first entry can correspond to the first page with a table of content, and each row of the table consists of links to subordinate pages of the hierarchy. ing. Returning to the calling process can occur after step 503.

After determining the appropriate first entry to the process, the derivative content may be recursively or recursively prepared to prepare the derivative content until the derivative content is prepared for all identified entries. Guidance corresponding to the entered entry is available.

According to various aspects of the invention, a model
Guidance prepared for the content can be applied for the currently requested primary content (step 30).
2). The process of determining whether to apply such guidance to primary content includes determining whether there is a sufficient correlation between the model description and the primary content description. The description can include a subset, summary, or reproduction of all or some of the described content. The reproduction may be in a markup language. The markup language used for description is
The markup language used to represent the content being described may be different. In a preferred embodiment, the description of the model content and the description of the primary content are provided in tagged elements in a markup language.

In accordance with various aspects of the present invention, guidance is captured, stored, and implemented using a markup language referred to herein as MASK. MASK is X
Matches ML and is therefore compatible with it. MAS
A partial list of elements of the K markup language is provided in Table 4.

[0077]

[Table 8]

[Table 9]

[Table 10]

[Table 11]

A description of the content to ensure particularly suitable guidance is captured, stored, and
It is used to analyze as a so-called frame surrounded by K tags <sklt> and </ sklt>. In accordance with various aspects of the invention, the description skeleton is based on the structural aspects of the content (eg, node hierarchy and table nesting).
And weaken the information conveyed by the content (eg, news or stock price matters). As a result, as described above in connection with process 412, the illustrative skeleton correlation is efficiently achieved by any of the conventional techniques of string comparison and alignment. The partial list of elements of the description skeleton is configured to describe the content expressed in the HTML markup language, Table 5
It is described in. The use of a single character (e.g., one byte) for structural features (e.g., the beginning of a table definition) provides for efficient creation, storage, and analysis in a compact description.

The PC skeleton creation process 410 can be said to be statistics 420 for each node to determine an appropriate description for each node. If a node contains a variety of features (eg, a mixture of text, numbers, and links), one description can be selected that corresponds to the preferred feature. The following description will be understood as indicating preferential features. For example, "_" mainly corresponds to a node having a link, even though text numbers may also be present.

[0080]

[Table 12]

As described above, in order to check in advance the correlation between the model content and the current primary content,
A description of the model content can be obtained as associated with the address (eg, steps 502 and 503) (step 504). The description of the model content is
It may be a tagged element in an OBJECT attribute consisting of text in markup language. The description of the primary content (or the primary content area) is described in process 4
Prepared as described above in connection with Step 10 (Step 5
04).

A skeleton of the current content description (eg, stored in temporary storage) and (eg, MA
SK Markup Language Guidance <sklt
>. . . . Conventional string alignment and comparison can be used to determine the extent of correlation with the previous content description skeleton (retrieved from the tagged portion of </ sklt>) (step 508). A conventional Smith-Waterman process can be utilized.

In an alternative embodiment, a type of transition model known as a Hidden Markov Model can be used. Progress models are specific content (for example, a set of structural features)
May appear at a particular location in the description skeleton, at a particular screen location, or reflect a probability that is logically or positionally related to other particular content. The progress model can be created based on one, or more preferably, many examples of the primary content. A second model of the current primary content is prepared to determine whether there is sufficient correlation between the current primary content and the progress model, and the progress model and the second model are compared to any conventional model. Can be compared.

If the degree of correlation is sufficient, the guidance associated with the previous content description is returned as appropriate guidance (steps 510 and 514). If not, the default guidance can be returned (steps 510 and 512). Logic or numerical thresholds and comparisons can be used in determining whether the correlation is sufficient. For example, if it is sufficient to find a match, a check of the binary logic that a match was found determines the result of the test (step 510). In an alternative embodiment, the weighted sum of the range of matches found (eg, allowing for intermittent mismatches in the correlation) can be compared to a threshold value. For example, if the weighted sum exceeds that number, it can be concluded that there is sufficient correlation. In another embodiment, the probabilities derived from the model are combined with any conventional method,
It is concluded that there is sufficient correlation. For example, if a match is found in a portion with a low probability of change, or if most mismatches are associated with a portion with a high probability of change, it may be concluded that there is sufficient correlation.

A complexity reduction process according to various aspects of the present invention includes some process of recognizing content for better navigation. The presented content is too complex to present in audio or limited presentation, so the content is reviewed to recognize its organization. Complex content is content that is difficult for an average user to understand or navigate when spoken, and has limited display area and / or resolution. Content that is difficult for an average user to understand or navigate when presented partially on a screen. Navigation involves understanding the presentation, recognizing the links, selecting the desired link, and directing access to information related to the desired link. Complexity can interfere with such navigation step or steps.

If the organization is such that regions in the content can be recognized, a further analysis is performed for each region, more or less independently. An area is a part of the content that has little contextual relevance to other parts of the content. Contextual relevance may be structural or aesthetic relevance. Structural differences include differences in presentation, for example, text presented in the first table may be text that is not in the table, text that is separated from the table, and text that is present in the second table. It may be completely dissociated from the text presented. Aesthetic differences include appearance,
Includes physical location or placement differences, for example,
If the content (eg, expressed in a markup language) specifies a first passage of text and links on a first background and a second passage of text and links on a second background, The relevance of the first and second passages is treated as weak by implication, and another region is defined for each passage. Many pages have headline information presented at the top of a large screen and legal notices and general information presented at the bottom. Blank spaces indicate segmentation. The relative physical location may indicate importance. After recognizing the differences that define the regions or the expected importance associated with each region, further analysis can be performed for each region in order of relative importance. A region to be further analyzed is selected based on the above criteria (step 524).

An analysis of the content to be presented is performed to determine whether the selected region should be further subdivided (eg, forming a hierarchical set of pages representing the contents of the region). You. For example, the complexity of the region can be determined by calculating a measure of complexity or complexity (step 526). If the content is expressed in a markup language, the measures may include the number of bytes in the markup language, the number of bytes of the content in the area, and the number of links in the area to represent the content of the area.

If the region is not determined to be too complex (step 528), the identifiable portion of the region is analyzed for relative importance (step 530), and
Derived content 440 can be configured. Derivative content can include content in a markup language that need not necessarily be the same as the markup language of primary content 400. Derivative content is ranked based on rank of importance (eg, from the upper right corner of the screen to an absolute position, relative to any desired position of the predicted screen, or relative to another part of the area). (Step 53).
2).

On the other hand, if the regions are determined to be too complex (eg, if one or more scales or weighted scales exceed the limit), each has the content of the selected region. A set of partitioned components of derivative content can be prepared (step 53).
4). The importance rank of each partitioned component can be ascertained as described in connection with step 530 (step 536). Further, the derivative content may be prepared to include a list of links to each partitioned component of the set. Any suitable summaries and annotations can be accompanied by links in the derivative content.
(Eg, annotation process 436 or primary content 4
Certain content identified for presentation (by 00) can be omitted from the set.

After the derivative content has been expanded in a preliminary form (steps 532 and 538), the font, position, background, and many other contents that can be controlled in a conventional manner via a style sheet are controlled. Any conventional style sheet (or cascading style sheet) can be applied to affect the property (step 540). The derivative content can be revised for the inline style (step 542), or it can reference a style sheet in a file or message that is stored separately from the derivative content.

The primary content can be used and analyzed,
Derived content for all piecewise components can be prepared. Alternatively, if a piecewise set of components has been identified to be prepared (step 534), the first component (or summary of the set)
Can be prepared in step 542, and the remaining components can be prepared as needed (eg, following a link). When presenting components of a set, to simplify navigation between the components of the set, or to indicate that the preparation of derivative content for another component is desired,
You can send or update traditional cookies. Derivative content for all regions can be prepared by steps 524-544 for all pages and all regions (step 544).

After step 544, control can return to the calling process. In a preferred embodiment, the annotated content is stored in the DOM as described above in connection with the primary content analysis process 406 and record 418. Record 418 is stored as DOM in memory. The complexity reduction process 408 is notified by the annotation process 436 of the PC node identifier that it is ready for further processing. The received PC node identifier identifies the node (corresponding to the area of step 524) to determine complexity. The children of the identified node may correspond to portions of the content (step 530) or may correspond to component pages of the set (step 534). Steps 524-538 can be accomplished by applying a set of rules to the identified nodes. The rules may be applied in any order, or may be organized as nested if-then-else (or case) statements to be applied in a fixed order. The actions taken based on the preferred set of rules are listed in Table 6.

In Table 6, the dominant node can be defined as a node having a high content ratio of the primary content. This ratio is the statistic applied to the top node of the DOM, for example:
It can be calculated using statistics (eg, num_bytes) applied to this node divided by (num_bytes). For example, the number of bytes of text content to be presented for a pair of tags at this node (e.g., from <form> to </ form>) is the number of bytes of text content to be presented as a whole on the page (e.g., <html> to </ html>) is dominant. If the ratio is 2/3 (ie, 67
%), The node can be considered a dominant node.

[0094]

[Table 13]

[Table 14]

A method for presenting derivative content in accordance with various aspects of the present invention includes the steps of identifying a style suitable for audio and / or visual presentation,
One or more of presenting the selected, annotated and reproduced content, or simultaneously capturing and analyzing user input to handle interrupt commands from the user during presentation. Any method including: Method 300 above
The method in the embodiment invoked from step 306 begins by identifying one or more styles suitable for the information to be presented. For example, FIG.
In the system described above, the audio browsing process 2 that receives a content (eg, PAGE2) in a markup language (eg, VoiceXML)
16 identifies styles from the content (e.g., inline or referenced) (step 60).
2) Alternatively, an appropriate default style may be specified. The default style is an indication of the type of audio device 202 that can be implied or specified in previous communications (eg, registration of the device 202 and its users with an authority to provide audio browsing services, or, in step 242, an audio browser. (By additional code or cookies that can access the server 123). The style information can be said to be a dictionary of pre-defined or user-defined terms with preferred pronunciation pronunciation instructions. For a particular term, the instruction can indicate that the term is to be spelled, rather than pronounced.

After a style has been specified (or applied) for application, two parallel execution paths can be supported. First, the utterance and / or display of the derivative content (step 604) is directed to the user's device (eg, 202).

Second, during the evolution of the presentation, inputs (eg, speech or keypad switch actuation) are accepted and accumulated for recognition (eg, speech recognition, or multi-key sequence recognition) ( Step 606). If the input is recognized as including a command (step 60)
8) The speech and / or display can be interrupted (step 610). An acknowledgment of the command can also be provided. The status of the command is noted (eg, when the command was initiated, or where it is presented (eg, TTS queue, or displayed content).

[0098] If the presentation is complete, if the interrupting command is not identified, and if the interrupted command and its status are noted, it is possible to return to the calling process. If there is an index of the command that was interrupted, that index and the status of the command can also be returned.

A message sequence obtained by a process operating on a system according to various aspects of the present invention may be a device having a visual area and / or a display with a limited resolution (referred to herein as a limited display device). ) Can be provided. The limited display user interface allows, among other things, a user to access information stored on the network (eg, navigating the Internet by following hypertext links). For example, the message sequence 700 of FIG.
The processes cooperate in requesting information and presenting the received information. For clarity,
The Internet and its conventional protocols perform the functions of network 130, information is requested by following links, and information is received in the form of web pages. The message sequence 700 may be a connection-oriented link or a connectionless link (eg, 17
1, 173 or 175) are already available for communication (formed as described above or in any conventional manner) and information with links (eg hypertext links) is limited in display It is assumed that the currently displayed device 702 is displayed.

The device 702 with limited display refers to any access device that provides a display with a smaller area or lower resolution, especially as compared to the conventional monitor of the workstation 111. Examples of devices whose display is limited include (a) display of only text, monochrome
Wireless device 112 capable of text and graphic display or pocket-size color display; (b) telephone 113 with video display panel with diagonal size less than 8 inches (10 cm); (c) (a) and (b) ) Includes a mobile phone 114 having a display device similar to that described in the above. Device 70 with limited display
2 assumes that it has minimal computing power sufficient to perform I / O process 706 for clarity. The I / O process 706, performed by the display limited device 702, receives user input requesting access to information addressed by the hypertext link and issues an appropriate request to the browsing process 708. (Step 730).

In an alternative embodiment, the limited display device 702 may have sufficient computing power to perform the browsing process 708 (eg, commercially available from Microsoft to operate on wireless devices). Internet Explorer). In that case, the browsing server 704 may be omitted with changes accompanying the message sequence 700.

The browsing process 708 includes a network 130 (eg, TCP / IP, WAP, HTTP for wireless devices and / or the Internet).
Etc.) may be involved in any of the conventional communication protocols associated therewith. Browsing process 7
08 can be executed by the browsing server 704. For devices with limited display of the type described above in connection with telephone 113 and mobile telephone 114, browsing process 708 may be performed by audio browsing server 123 instead of browser server 704. As in the above case, the browsing server 70
4. The functions of the transcoder proxy server 125 and the ISP server 124 may be performed by any number of computers (including one) at any number of physical locations.

MASK database 230, browsing process 708, transcoding process 21
8, the functions performed by the routing process 222 and the serve process 224 are URL3, URL3.
4, messages corresponding to PAGE3 and PAGE4 may correspond to the functions described above. (That is, Steps 746 to 760 in FIG. 7 correspond to Step 2 in FIG.
66-280. URL 3 also includes an identifier for browsing process 708 so that transcoding process 218 can prepare PAGE 5 in a form suitable for use by browsing process 708 (instead of PAGE 4 of step 280). Good.

When the browsing process receives the information (eg, PAGE 5) from the transcoding process 218 (step 760), it provides an appropriate display corresponding to the requested information to the limited display 702. (For example, the message PR
ESENTATION) (step 73)
0). PAGE4 is a markup language (eg, HTT
P, XML, or WML) page. PRES
The ENTATION is delivered via a priority protocol adapted to the device 702 whose display is limited (eg, conforming to the specifications of the manufacturer of the device 702). Adjustment of the display size and / or resolution of the device 702 with limited display is performed in the database 2
The transcoding process 218 and / or the browsing process 704 dictated by the guidance from 30. Preferably, the identification of the device 702 with limited display and / or the browsing process 704 as received at the URL 3 (eg, as described above with respect to step 246)
As used by the transcoding process 218 to obtain guidance from the database 230,
PAGE 5 only needs to apply one or more conventional style sheets to provide the appropriate PRESENTATION. In other words, if the requested page exceeds the complexity appropriate for the limited display device 702, PAGE 5 may include a portion of the set of pages as described above.

In accordance with various aspects of the present invention, a system and method for preparing guidance is provided for providing a graphical user interface with which a user can retrieve information from a computer network and provide guidance related to the retrieved information. Provide an interface to the user.
Thus, as described above, the guidance can be used for a voice user interface or a user interface with limited display. For example, if the computer network includes the Internet, a user may search for web pages and run a browsing process to provide guidance of the type described above. The user can perform operations (eg, mouse clicks) in a graphical environment to instruct the preparation of guidance by automatic text creation. The generated text may be in a markup language, such as the MASK markup language, as described above.

According to various aspects of the present invention, the method of providing guidance includes selecting, among other things, a portion or area of information.
Includes any process that includes annotating the selected information and / or creating a description of the information for use in associating guidance with the selected information (eg, a skeleton of the description). For example, the message sequence 800 of FIG. 8 can be accomplished by the workstation 111, the transcoder proxy server 125, and any of the site servers 142. These devices have been described above, including a serve process 226 performed by any of the site servers 142. Browsing performed by workstation 111
Process 832 may include any information that requests and presents a suitable program for network 130. If network 130 includes the Internet, browsing process 832 may include any conventional Internet browser (eg, Internet Explorer marketed by Microsoft Corporation). Transcoder proxy server 12
5, the MASK edit process 834
The configuration can be made using any conventional programming technique suitable for use on network 130. For example, the MASK editing process 834 may include, for example, C ++, PERL, CGI APIs, JA, to name a few.
VA (registered trademark), Java (registered trademark) Script
It can be deployed using known languages and interfaces, including t and Xwindows.

The guidance specified during the editing session defines what information is presented and how it is presented in the derivative content. Derivative content applies the guidance specified during the editing session,
Derived from primary content (eg, a model page is the prototype for it).

The message sequence 800 will now be described assuming that the network 130 includes the Internet. In response to user input, the browsing process requests a page from MASK edit process 834 (step 802). The user can know the address of the requested page, or the user is directed to the appropriate page by a hypertext link on another page that the user knows. The URL of the requested page may include an index of the model page to be edited (in any conventional manner).

Upon receiving a request from the browsing process 832, the MASK editing process 834 requests a model page (or default page) from any of the site servers 142 (step 804). The serve process 226 returns the desired model page in any conventional manner (eg, in a markup language with animation, streaming audio and / or video) (step 806).

The MASK edit process 834 prepares the presentation based on a portion of the model page (step 80).
8), send the presentation (eg, EDIT PAGE) to the browsing process 832; Presentation can be reduced in complexity in any such manner (eg, as dictated by guidance from database 230), or in any conventional manner. The presentation can be made using the audio user interface described above or a user interface with limited display, in which case the predefined guidance is associated with EDIT PAGE. Preferably, efficient deployment of guidance for model content is obtained by the presentation, including one or more elements, as schematically illustrated in screen 900 of FIG. 9 and listed in Table 7. .

[0111]

[Table 15]

[Table 16]

[Table 17]

[Table 18]

When the user enters a text box (for example, 9
06, 908, or 910), or complete the entry of the text (e.g., 912-9).
22) or changing the assertion of one or more check boxes, a message describing the action (s) performed may be sent to the MASK edit process 834 (step 810). The message may be in the form of a URL with parameters. The MASK edit process 834 responds with an updated edit page presenting a panel 902 that has been cleared and is ready for the next operation, and a panel 904 that will be updated to show the effect of any accumulated annotations. Yes (Step 8
12). These two operations (steps 810 and 8)
12) is repeated a desired number of times. Button SA by user
Browsing process 8 in response to VE assertion
32 may send a message indicating that editing is desired to end (step 814). The MASK editing process 834 then sends the guidance to the MASK database 230 (step 816), and
An acknowledgment (eg, a predefined page) may be sent to process 832 (step 818).

Before the end, MASK editing process 834
Can analyze and prompt the user to specify information that will be used to access guidance resulting from the editing. For example, as mentioned above, the user may have NAME, DESCRIPTION, USE
R_ID, MODEL_ADDRESS, APPLIE
S_TO_ADDRESS, HAS_REGULAR_
You are prompted to provide information corresponding to one or more of EXPRESSION or IS_FIRST. The edit process 834 can provide further analysis or default values.

The guidance can be stored as one or more records in database 230 in any conventional manner. Guidance may be indexed based on the URL of the model content. Since the guidance can be indexed by a URL that constitutes a regular expression derived from the URL of the model content, the guidance shall be applicable to all primary content URLs that match the regular expression. Is displayed. Guidance may be stored or accessed in a hierarchical manner that maps parts of the URL to a directory (or holder) with subdirectories (or subholders). The method for accessing guidance includes the following steps: 1. Search for the URL exactly as presented (eg PC URL in FIG. 4), stop if found,
If not found, continue. 2. Truncate the last (child) part of the URL. For example,
“Http://www.news.com/US/2
000-OCT-30. “htra” means “http: //
www. news. com / US / ". If a censored URL is found, stop; otherwise, continue. 3. Search for siblings of censored URL in step 2,
If found, use guidance related to siblings. If not found, continue the search. In the example of step 2, this step 3 is executed at "http: // www.
w. news. com / US / [. ] ^* "4. U, keeping only the schema and initial domain specification
Terminate RL. If an aborted URL is found, stop. If not found, no guidance is available. For example, derivative content 440 is prepared by the cooperation of analysis process 406 and complexity reduction process 408.

The following sequence of user input, also set forth in Table 8, creates guidance for model content (eg, selected and annotated content) represented by the schematic representation of FIG. 10A. Will do. "Click" means that the left mouse button is pressed and released when the mouse pointer is positioned over the displayed feature.

[0116] 1. Click box 950 to indicate that the entire table 950 will undergo subsequent actions. 2. Click the button 914SELECTION,
Let table 952 appear in the derivative content. 3. Click box 980 to indicate that one cell of table 950 will undergo a subsequent operation. Box 9
Click button 916 REMOVE to omit 80 from the derivative content for table 952. 4. Click in boxes 956 and 958 and select PRE
“FOR” is input into the AMBLE box 908. In addition, clicking the button 912 DESCRIPTION presents the word "FOR" (e.g., spoken) before the remaining content in each displayed cell. 5. Click in boxes 968 and 970 and select PRE
Enter “CHANGE IS” in the AMBLE box 908. In addition, button 912 DESCRIPTION
Clicking on N presents the word "CHANGE IS" before the remaining content in each displayed cell. 6. Click in boxes 974 and 976 and select PRE
Enter "AT" in the AMBLE box 908, and
“POINTS” is input into the AMBLE box 910. Further, clicking on button 912 DESCRIPTION presents the word "AT" before the remaining content in each displayed cell and the word "POINTS" after the remaining content. 7. Assert only box 954, then button 92
0 Click ENFORCE. A “Keymask =” argument is added to the derived node of table 952 in the description section and / or the selection section. 8. Click the button 921 SKELTON. A skeleton of the model content is created. An "Sp =" argument is added to each node in the description and / or selection section. 9. Clicking the button 922 ends the editing, and saves the text in the box 906 as guidance. The guidance may also include data corresponding to the text in box 906. For example, skeleton is record 415
And can be stored in a format including the relationship between the above-described types of node names and skeleton characters.

[0117]

[Table 19]

[Table 20]

As a result of the above command sequence, if the primary content similar to the model content of panel 904 is transcoded, only table 952 will appear in the derivative content. Table 952 is presented with rows 1082 and 1084 revised to make the speech utterance more understandable. For example, the utterance based on FIG. 10A includes “Market Snapshot.
And minus 49 and 6400 at 5000 points
is. For Natsudag, they are minus 101 and 4000 at 5042 and 100 points. Is included.
The pronunciation of NASDAQ can be specified in the dictionary as described above.

If the set of pages is, or will be, prepared for display via an audio user interface or a limited display user interface as desired, the transcoder 218 Can include additional links on each page. For example, FIG. 10B shows a display formed based on the guidance as described above in connection with FIGS. 9 and 10A, and showing derivative content from the primary content from a different date than the model content. Indicates the display contents of the device whose number is limited. Display 109
0 includes a content 1091, an auxiliary link 1092, and a switch link 1094. Auxiliary link 1
092 may be other pages (for news content, weather content, and sports content) derived from the same primary content (eg, by a summary as described above or by the user's provisions in the guidance for each auxiliary page). Supply access to In addition, following link 1094 facilitates the transfer of session control as described below with reference to FIG.
For audio utterances, the auxiliary link may include uttered control functions, such as a menu of items as described above with reference to Table 2.

On the limited display screen of FIG. 10B, about 6 lines each having a length of about 45 characters are displayed. If the length of the displayed screen line is less than 45 characters, the illustrated line can be duplicated with additional lines. Devices with limited display have a buffer to hold more lines than fit on the display screen, and scroll controls operated by the user to effectively move different parts of the buffer to the display screen for presentation It has.

[0120] Certain guidance can be applied to primary content to create derivative content. As described above, model content can be used to develop predetermined guidance. The prescribed guidance may be, for example, M
A description of the model content, such as a description in the ASK markup language, can be included. In addition, the description of any primary content may (a) specify whether certain pre-defined guidance associated with similar model content should be used with the primary content, and (b) derivative Prepared and used to prepare content. Initial use can be achieved by comparing the primary content description with the model content description. A second use can be achieved by aligning the model content description with the primary content description.

For example, in the method 304 of FIG. 4, the preparation of derivative content is performed by the transcoder proxy server 125 as part of the transcoding process 218.
Can be done by To prepare derivative content based on the primary content, various aspects of the invention may be used, for example, as described above for the UR of the primary content.
With reference to L and a URL including a regular expression, the position of the skeleton of the description of the appropriate model content is found.

After the primary content has been received, a skeleton of the primary content description can be created. For example, primary content substantially similar to the model content schematically presented in panel 906 of FIG. 9 is presented in Table 9 in HTML. Standard primary content is significantly larger in volume and complexity. Panel 906 is prepared for readability, especially by reducing actual news, adding table rows, and adding check boxes to select various features for convenience in preparing guidance. Note that you can

[0124]

[Table 21]

Using the MASK markup language, Table 9
The description of the primary content is shown in Table 10. White space has been added to make the comparison clearer,
In general, they do not appear in the description framework.

[0126]

[Table 22]

The MASK markup language uses non-structural elements (eg, table cell content, text, numbers, links,
Graphics, etc.). Numbered nodes do not explicitly appear in the primary content markup language, but nesting and nesting of tags used for content
These are clear from the level analysis. A pair of tags surrounding the content means a tree node. The root and all nodes at the same level are numbered from left to right, starting with zero. Examples of nodes are shown in Table 11.

[0128]

[Table 23]

Using the nodes numbered as shown in Table 11, the unstructured content “Dickson” is converted to derivative content using the node number “0-0-0-1-1” as a reference. And could be copied. Since the numbering of the nodes of the primary content may change (e.g., a new table is added to the body), the content copied to the derivative content will have different structural and unstructured primary content. It can be specified based on the features. For example, in the model content shown in FIG. 10A, to find the location of rows 982, 1082 where Dow Jones' average stock value may be found (instead of the word stock market snapshot) The word "DOW" may be used. Using the reference to row 982, other rows, and the overall node name of table 952, can be determined (eg, by process 428) in the model content used with the primary content. Node names with relative references are easier to align with later (eg, another) primary content.

According to various aspects of the invention, reference to content to be copied to derivative content is made with reference to one or more anchors. By aligning some or all of the skeletons of the description as described above, or any one or more of the elements (e.g., the text The string "DO
An anchor can be found by aligning W ", a passage of the predicted word count, a link count, a link with the predicted word count, etc.).

Step 8 of Table 8 is an example of an editing session and guidance including the model content described above with reference to FIG. 10A. Each anchor is <node><
Note that it is specified by the specification between the / node> tags. In this example, a plurality of key masks are defined in addition to the node number and the skeleton position. The transcoding process 218 locates the desired content for inclusion in derivative content based on either one or more anchor specifications.

The derivative content is workstation 1
Presentation, audio user interface and audio device 202 through 11 browsing processes (see Table 2)
(See Table 13), or a presentation using a limited display user interface and a limited display device 702 (Table 14). The derivative content shown in Tables 11-13 is the primary content of Table 8 and Table 8, Step 8
10A, and thus includes only the features of table 952 in FIG. 10A. For example, the cell 980 is omitted.

[0133]

[Table 24]

The voice user interface uses male and female voices to show the difference between information and links. The link is announced in a female voice. Following the link, the announcement of the content may reproduce the same word for orientation to the user, but the reproduction occurs in a male voice. For example, the first time a "form 1" is encountered as a link, a female voice will announce this. If the user says "Format 1" to follow the link, a male voice will acknowledge that the link has been followed and the title of "Format 1" will be given before the Format 1 content is published. "Can be announced.

[0135]

[Table 25]

[0136]

[Table 26]

According to various aspects of the present invention, a user having multiple access devices (or devices suitable for use with multiple user interfaces) may have more than one access device at any time, sequentially or in parallel. Access devices (or user interfaces) can be used. The user enters via any of the above user interfaces (eg, a workstation GUI, a voice user interface, or a user interface with limited display) and from that point on one or more of these interfaces. Perform system operations. For example,
The user initiates a session to access the Internet via the user interface of the wireless device 112 or 702 and the device with limited display, and informs the browsing process 708 that the session is continuing in audio. And possibly to an audio user interface (e.g., audio browser 216 described above) to indicate that the session is to be resumed on a device with limited display. In an alternative embodiment, the session switching is performed by the workstation 111 GUI.
Timely control by 832 may be included.

During the transfer of control, these access device or devices may control the session exclusively or non-exclusively (eg, input at any user interface may be browsed). , And the output is provided to all access devices in parallel). The output may be provided exclusively to the interface used exclusively or may be provided to two or more interfaces as instructed by the user.

For example, from a user interface of a device whose display is limited to a voice user interface, or again a user of a device whose display is limited.
To facilitate communicating control of the session exclusively to the interface, the message sequence 1100 of FIG. 11 can be utilized. In this sequence, 1
The human operator has the ability to support the voice access device 202, the limited display device 702, and the browsing process 1105, the calling process 1108, and I (to the extent not implemented in the circuit).
Use a web phone 1102 that includes an / O process 206. The web phone 1102 initially establishes the link 171, the wireless gateway 121, the link 172, the ISP server 12 as the wireless device 112 of FIG.
5 and the Internet 130 by link 178
Connected to. In an alternative embodiment, the web phone 11
02 is a link 171, wireless gateway 121, link 174, transcoder proxy server 125, and link 178 as wireless device 112.
Is connected to the Internet 130. web·
The phone has been shown to be already involved in a session with the voice / LDD browsing process 1106 on the browser server 1104. The browser server 1104 of the alternative embodiment includes the functionality described above in connection with the voice browser server 123 and / or the browser server 704. The audio / LDD browsing process 1106 supports both an audio user interface, as described above, and a user interface for devices with limited display (LDD).

The browsing process 1105 may be a user input (eg, an arrow key or pen event).
Respond to The user can navigate the Internet in the conventional way of selecting links using user input. At any time, the user may wish to continue the session via the voice user interface. To do so, the user can use the audio / LDD browsing process 1106 to provide links (eg, FIG. 10B for current content in audio)
(Or link 1096 of FIG. 10B to an Internet wireless site) (step 1120). The link is TO_AUDIO
A URL corresponding to the command can be included. Alternatively, the device whose display is limited is the command TO_AU.
DIO to audio / LDD browsing process 1106
May have software, or software, or receive user input therefor. The command is URL,
For example, it can match the URL of the browser server 1104. The browsing process 1106 includes enough Web resources to comply with the approval for the voice session.
By obtaining information from the phone 1102 and configuration information to establish an appropriate voice connection, the TO_AU
Responds to DIO command. Such information is available on the web
From phone 1102 or browsing process 1
The registration data obtained by the browsing process 1106 when the user first starts a session with 106 may be obtained via one or more cookies or URL parameters. As described above,
A login scenario can also be implemented by the browsing process 1106. TO_AUDIO command is UR
If L, the parameters of the URL may specify a particular page to be served in audio (eg, the same information as in the current page, another page based on the same primary content, Internet wireless site, List of audio sites). The identifier of a particular page (eg, URL9) is the user information 1110 (eg, username,
Caller ID, calling telephone number, or registration information) (step 1121).

After the switch to the voice user interface has been approved and confirmed to be appropriate, the browsing process 1106 can notify the user of an appropriate acknowledgment (step 1122). The acknowledgment may include a link to establish a voice session and a destination telephone number. In one embodiment, the telephone number is W
Part of a link in the ML markup language (eg, <ahref "wtai: // WP / mc; 1aaa"
eennnnnn "> TO AUDIO </a>, where wtai is a type of schema known as wireless telephone application interface;" wp / mc "is WT
Identify the APublic library and the command "mc" to place the call; "aaaeennnnnn" is a 10 digit U.S. phone number).

The user can access the web server 1102 (and the calling process 1108) from the browser server 1104.
A call can be made (step 1124) to (and the browsing process 1106) the link can be followed (step 1123). For example, the browsing process 1106 may send an applet to the browsing process 1105 with an acknowledgment (step 1122). The browsing process 1105 can then initiate a telephone call from the web phone 1102 to the browser server 1104 and provide a user identifier (step 1125). The browser process 1106
The user can be authenticated using conventional caller ID technology (step 1125).

[0143] In an alternative embodiment, the command TO _ AUD
The IO causes the browsing process 1105 to prepare for the incoming call, and the voice / LDD browsing process 110 (with sufficient approval to proceed)
6 is placed to make a call to web phone 1102.

Regardless of whether the browsing process 1106 or the browsing process 1105 initiates a telephone call, the web phone 110
2 via link 175, cellular gateway 122, and link 176, browser server 1104
1 is connected as the mobile phone 114 of FIG. Voice browser server 123 is connected to Internet 130 via link 177, transcoder proxy server 125, and link 178.

[0145] As indicated in the TO _ AUDIO command by the user, or in any conventional manner, and the publication of the voice of appropriate initial page audio / LDD browsing process performed by the audio device 202 110
6 to the I / O process 206. For example,
The user of the device 702 whose display is restricted is supplied with the URL 9 stored together with the user information 1110. To recreate the URL 9 identification of the desired content, the voice / LDD browsing process 1106 requires the caller I
D or other user information is available (step 1126). The browser process 1106 transcodes the primary content of the URL 9 as described above, or obtains the audio content flowing from the Internet 130, and sends it to the I / O process 206 as the RECITAL 9 (step 1127). A user of the limited-display device 702 may wish to listen to transcoded primary content based on the above guidance. In this case, the browsing process 1106
Provides the functionality of the speech browser server 123 (eg, speech recognition for navigation commands) and cooperates with the transcoder proxy server 125.

[0146] A user at any time, browsing commands (TO _ LDD) to continue a session in the device 702 of the display which is limited with respect to I / O process 206
It can be instructed to transport (step 1128) to process 1106. The command may include a DTMF signal or a voice command by the user, such as "Send page", or any other suitable command as shown in Table 2. Browsing process 110
6 responds (step 1130) with, for example, appropriate derivative content (PAGE 10) as described above in connection with step 780. TO _ LDD command is recognized by the speech recognition software to notify other processes browsing process 1106 in any conventional manner so as to switch to a user interface device for a display that is limited from the voice user interface Can be

For the sake of clarity and ease of description, some specific embodiments of the present invention have been described. It is intended that the scope of the invention be defined only by the claims. The description is not intended to be exhaustive, and is not intended to limit the invention to the form disclosed. In view of the disclosure and embodiments of the present invention, other embodiments will be apparent to those of ordinary skill in the art to which the present invention pertains.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a system according to various aspects of the invention.

FIG. 2 is a message sequence diagram of an embodiment for providing a voice user interface of the system of FIG.

FIG. 3 is a dataflow diagram of a method for supporting various user interface functions of the system of FIG. 1;

FIG. 4 is a data flow diagram of a process for preparing derivative content from primary content and guidance in accordance with various aspects of the invention.

FIG. 5A is a data flow diagram of a method for discovering guidance implemented in the process of FIG.

FIG. 5B is a data flow diagram of a method for reducing the complexity of presented content performed in the process of FIG.

FIG. 6 is a data flow diagram for presenting derivative content in the method of FIG. 3;

FIG. 7 is a message sequence diagram of an embodiment for supporting the limited display device of FIG. 1;

8 is a message sequence diagram of an embodiment for providing a guidance editor in the system of FIG.

FIG. 9 is a graphical user interface screen layout for editing guidance based on FIG. 8;

FIG. 10 is a diagram showing a part of the screen layout of FIG. 9 after a guidance revision is instructed.

FIG. 11 is a message sequence diagram of an embodiment of a method for supporting a device having both an audio device and a device with limited display.

[Explanation of symbols]

110 access device, 111 workstation,
112 wireless device, 113 telephone, 114 mobile phone, 120 gateway device, 121 wireless gateway, 122 cellular gateway, 1
23 Voice Browser Server, 124 Internet Service Provider Server, 125 Transcoder Proxy Server, 130 Network, 14
0 Network server, 141 Home site server, 142 Any site server.

 ──────────────────────────────────────────────────続 き Continued on the front page (71) Applicant 500466566 128 Worcester Loop, Los Gatos, California 95030, U.S.A. S. A. (72) Inventor G-One Saltair Avenue 1409, Los Angeles, California, United States of America, 1409, # 302 (72) Inventor Jin Chan United States of America, California, Pasadena, South Allen Avenue 159, Apartment 212 F term (reference) 5B082 GA02 HA05 HA08 JA01 5E501 AA02 AB15 BA03 CC17 EA34 EB11 FA13 FA14 FA22 FA32

Claims (55)

[Claims] 1. A method for transcoding information provided in a first markup language, comprising: selecting a portion of the information; determining a complexity of the selected portion; Creating a hypertext link to access the subset of information and the subset of information if the complexity of the selected portion is beyond a limit; Providing said hypertext link and at least one of said subsets and providing annotations associated therewith. 2. The method according to claim 1, wherein the step of determining the complexity comprises: determining a number of words, a number of words, a number of tables, a number of hypertext links, and a number of words in the hypertext links. The method of claim 1, comprising calculating a quantity and determining the complexity based on whether the statistic is out of bounds. 3. The method of claim 1, further comprising repeating the steps of selecting, determining, and creating for a plurality of pieces of the information, wherein the created hypertext links are listed. . 4. The step of determining further comprises the step of determining the importance of the selected portion, wherein the hypertext link is based on the importance of each selected portion. 4. The method according to claim 3, wherein the method is enumerated. 5. The repeating step is performed for a portion of the information accessible via a particular hypertext link after a user navigates the particular hypertext link via a list.
The method of claim 3. 6. A server for speech browsing and transcoding, comprising a method indicator according to claim 1. 7. The method of claim 1, wherein said providing step is performed via the Internet. 8. A method for transcoding information provided in a first markup language and associated with a first URL, the method comprising: providing guidance including a first description and an annotation to at least one of the first URLs. Obtaining a second description of at least a first portion of the information; determining a correlation between the first description and the second description; Providing the information and at least a second portion of the annotation if a limit has been exceeded. 9. The method of claim 8, wherein the second description comprises a key word. 10. The method of claim 8, wherein the second description includes a string of characters corresponding to the first markup language tag. 11. The method of claim 1, wherein the first markup language includes a meta tag, and wherein selecting the portion includes analyzing the meta tag to recognize a region to select. 12. The area is recognized based on whether any part of the information is in a table.
2. The method according to 1. 13. The area wherein the first portion of any of the information is identical to the second portion of any of the information with respect to at least one of appearance, placement, background, and relative screen position. The method of claim 11, wherein the recognition is based on whether or not there is an error. 14. The step of selecting the part,
The method of claim 11, comprising determining a respective rank of relative importance for each of a plurality of recognized regions, wherein the regions are selected according to the respective ranks. 15. The method of claim 14, wherein the region is further selected depending on whether any part of the information is in a table. 16. The method of claim 16, wherein the first portion of any of the information is associated with a second portion of any of the information with respect to at least one of an appearance, an arrangement, a background, and a relative screen position. 15. The method of claim 14, wherein the method is selected based on not being the same. 17. A method for communicating between a first server and an access device, the method comprising: receiving a request for information including a first address from the access device; The server communicates over a network and receives the provided data obtained from the second server at the first server; and accesses the guidance for transcoding at the first server. And wherein said guidance is associated with at least a portion of said first address; and transcoding said data based on said guidance to provide a user with a limited voice user interface and display. Providing derivative information according to at least one of the interfaces and at least an audio device and limited display Method comprising the steps of: transmitting said derived information to the access device for presentation based on one. 18. The guidance includes instructions to identify at least one feature of the derivative information, wherein the feature causes the derivative information to be presented in audio, or displays the derivative information, volume, conversation speed, Whether to present in a conversation pitch (interval) or direction, to suppress output of at least a part of the derivative information, to spell at least a part of the derivative information, at least a part of the derivative information 18. The method of claim 17, including an indication of at least one of: including speech font text, and including speech in at least a portion of the derivative information. 19. The data is received as expressed in a first markup language, and the guidance is accessed as expressed in a second markup language different from the first markup language. 19. The method of claim 18, wherein the method is performed. 20. The first address includes a URL,
The method according to claim 17. 21. The data is obtained from the second server based on a second address, wherein the second address includes an address of the first server and an address of the second server. Item 18. The method according to Item 17. 22. The method according to claim 17, wherein the communication between the first server and the second server is based on TCP / IP. 23. The data includes the guidance,
The method according to claim 17. 24. The method of claim 17, wherein the data includes a reference to the guidance. 25. Accessing the guidance further includes accessing a user account associated with a user of the access device to obtain user data, and accessing the guidance based on the user data. A method according to claim 17 ,. 26. The method of claim 25, wherein the user data includes at least one indicator of the type of access device and a user-defined preference, and wherein the guidance is accessed based on the indicator. 27. The step of accessing guidance includes: determining a second address based on the first address and at least a part of a regular expression; and accessing the guidance based on the second address. 18. The method of claim 17, comprising steps. 28. The method of claim 17, wherein the guidance is stored based on an attribute, wherein accessing the guidance further comprises comparing at least a portion of the attribute with at least a portion of the first address. The method described in. 29. The attribute according to claim 2, wherein the attribute comprises a regular expression.
9. The method according to 8. 30. The method of claim 28, wherein the guidance and the attributes are related for accessing based on a light directory access protocol. 31. The method of claim 17, wherein the guidance is stored in association with model content, wherein the guidance is accessed based on a correlation between at least a portion of the model content and at least a portion of the data. the method of. 32. The first address includes a scheme and a plurality of area designations arranged in a first character string, wherein guidance is provided for the entire first character string and a rightmost end of the first character string. A second character string derived from the first character string after truncation of at least one region specification from and a third character specifying at least one sibling node of a node associated with the second character string 18. The method of claim 17, wherein the method is accessed based on the string and at least one of a fourth character string derived from the first character string by truncation of all but one of the plurality of region designations. 33. The method of claim 32, wherein the accessing comprises a series of access attempts in the order listed in claim 32. 34. The access provides a plurality of candidates for guidance, and a respective model / model based on each candidate for guidance.
33. The method of claim 32, wherein content is accessed and a selected guidance is selected based on a comparison of the data description with a description of the respective model content. 35. The method of claim 26, wherein the access provides a plurality of candidates for guidance, wherein a respective skeleton of the model content is accessed based on each candidate of the guidance, 33. The method of claim 32, wherein the selected guidance is selected based on the comparison. 36. The data includes a meta tag of a first markup language, and the step of creating the derivative information includes excluding a part of the data from the derivative information with reference to the meta tag. 18. The method of claim 17, comprising determining at least one of the following steps: including an annotation portion of the data in the derivative information; and forming a summary of at least a portion of the data in the derivative information. Method. 37. The method of claim 36, wherein the summary includes a link to further derivative information. 38. The method of claim 37, wherein the further derivative information is created after indicating a link to be followed by a user of the access device. 39. The data includes content in a first markup language and a meta tag, the step of creating the derivative information includes: determining a plurality of node identifiers; Associating a node identifier with a hash version of a portion of the content; and outputting the selected content as derivative information, wherein the selecting step comprises: Positioning a specific content with reference to the specified specific node identifier as a reference, and positioning a specific content with reference to a specific hash version indicated by the guidance. The method according to claim 17. 40. The data comprises content in a first markup language and a meta tag, the step of creating the derivative information comprises: determining a first plurality of node identifiers; Associating each of the node identifiers with a portion of the content; and a second quantitative description describing a respective portion of the content associated with the first plurality of respective node identifiers. Determining a quantitative description; and accessing the first plurality of node identifier components to access a particular node identifier associated with the particular quantitative description; Associating and outputting the selected content as derivative information, where the selecting step includes a specific quantitative description and Locating the found node identifier based on a comparison with the limits indicated by the guidance, and using the found node identifier as a reference to use the content for use as the selected content. Accessing a particular part of the method. 41. The data includes data content and a meta tag in a first markup language, the guidance includes a description of the model content, and further includes a description of the model content according to a specific model node identifier. Analyzing the description of the model content to provide a first plurality of model node identifiers each associated with a respective portion of the description of the model content to access respective portions of the description. Analyzing the data with reference to the meta tag to access respective portions of the description of the data according to a particular data node identifier, and providing a description of the data description and each of the description of the data. Providing a second plurality of data node identifiers each associated with the portion; Analyzing the data to the tag to reference standard,
A third wherein each association is between a respective data node identifier and a respective portion of the data content.
Providing a plurality of associations, wherein each association is between a respective model node identifier and a respective data node identifier, each association being with a particular portion of the description of the model. Forming a fourth plurality of associations formed by comparison with a particular portion of the data description, the step of creating the derivative information comprising: outputting the derivative information as derivative information. Selecting content, wherein the selecting step comprises: locating a cross-reference table based on the association with the fourth plurality of associations; and determining a found node identifier based on the cross-reference table. Locating; identifying the content for outputting with reference to the found node identifier and the third plurality of associations; Comprising the steps of: accessing method according to claim 17. 42. The guidance comprises a skeleton, further comprising: creating a first set of records to associate a portion of the skeleton with a skeleton node identifier; and for associating a portion of the data with a data node identifier. Creating a second group of records, and creating a third group of records to associate a skeleton node identifier with a data node identifier, wherein the creating the derivative information comprises: Selecting a portion of the data for output as derivative information, wherein the selecting includes identifying the data with reference to the third group of records and the second group of records. 18. The method of claim 17, comprising accessing a portion. 43. For each particular record of the second group, determining a respective quantitative description of each of the content items associated with the particular record; Associating with each of the data node identifiers associated with the records of the above, wherein said selecting further comprises: determining a plurality of data node identifier candidates based on said guidance and said third group of records. Selecting one of the plurality of data node identifier candidates with reference to at least one respective quantitative description associated with at least one of the plurality of candidate references; Accessing said particular portion of said data as a reference. 44. For each particular record of the second group, determining a respective hash version of a respective portion of the data associated with the particular record; and Associating with a respective data node identifier associated with the particular record, the selecting step further comprises: selecting a plurality of data node identifier candidates based on the guidance and the third group of records. Determining; selecting one of the plurality of candidates with reference to at least one respective hash version associated with at least one of the plurality of reference candidates; further referring to the candidates Accessing said particular portion of said data on a basis. The method according to 2. 45. The step of selecting further comprises: determining a plurality of data node identifier candidates based on the guidance and the third group of records; each candidate model node identifier and the data
Determining a plurality of respective scores determined based on the node identifier; and outputting based on a candidate data node identifier selected based on at least one of a maximum and a minimum of the plurality of scores. Selecting a portion of the data to be processed. 46. Each score has the following formula: Is calculated based on, but, C _n is the candidate with scores S _n, each of the C has a level of P to the node name, M is a model node having a level of Q to the node name 46. The method of claim 45, wherein K is a constant and "min ()" returns to the minimum of its argument. 47. A transcoding step comprising: constructing, based on the data, a document purpose model including a data node identifier having a corresponding respective data node content; and translating each data node identifier. Associating each data node content with a data node statistic determined by reference; constructing a first description skeleton according to the document objective model; and a second description of the model content Obtaining a skeleton of the second skeleton; determining a plurality of model node identifiers corresponding to the second skeleton; and determining a model node identifier based on the alignment of the first and second skeletons. Associating with a data node identifier; Determining a complement; and accessing a particular portion of the data for output as derivative information based on a candidate data node identifier selected based on at least one data node statistic. 18. The method of claim 17, comprising: 48. The derivative information includes: determining an exclusion of the portion of the data from the derivative information including an annotated portion of the data in the derivative information; 48. The method of claim 47, created by at least one of the steps of creating at least a portion of the summary. 49. The derivative information determines a dominant node based on at least one data node statistic, and a set of dominant nodes, a summary of the dominant nodes, and a portion of the dominant nodes. At least one of a set of links
50. The method of claim 47, wherein the step is formed by forming one.
The method described in. 50. The data node statistic is a count of bytes in the text of the associated node content, a count of words in the text of the associated node content, a count of a table in the associated node content. ,
48. The method of claim 47, comprising at least one of: a count of hypertext links in related node content, and a count of words in hypertext links in related node content. 51. The plurality of data node identifier candidates are determined with reference to a part of a skeleton.
7. The method according to 7. 52. The method of claim 47, wherein the first description skeleton includes at least one of a table meta tag index, a table row meta tag index, a table cell meta tag, and a form meta tag index. The described method. 53. A skeleton of the first description is an index indicating any one line of text, an index indicating one line of any text exceeding a predetermined length, an index indicating any numeric string, any one of An indicator representing a hypertext link,
50. The method of claim 47, comprising at least one of: and an indicator representing a hypertext link that exceeds a predetermined number of words. 54. A communication system for performing the method of claim 17. 55. The system of claim 54, comprising a plurality of servers connected for data communication.