Abstract: Example methods for finding a text reading order in a document are described in which text zones are determined, the text zones are clustered using semantic measure and correlation and a reading order is found within each of the clusters.

Abstract: An automated document processing system is configured to normalize zones obtained from a document, and to extract articles from the normalized zones. In one configuration, the system receives at least one zone from the document, and applies at least one zone-breaking factor, thereby creating normalized sub-zones within which text lines are consistent with the at least one zone-breaking factor. The normalized sub-zones may be evaluated to obtain a reading order. Adjacent sub-zones are joined if text similarity exceeds a threshold value. Weakly joined sub-zones are separated where indicated by a topic vectors analysis of the weakly joined sub-zones.

Abstract: Dialog management within a call handling system includes monitoring a dialog between a contact and an operator. A first dialog attribute confidence score is generated based on an acoustical analysis of the dialog, and a second dialog attribute confidence score is generated based on a keyword analysis of the dialog. The first and second dialog attribute scores are combined, and a rule is effected in response to a value of the combined dialog attribute score.

Abstract: A system, method, computer-readable medium, and computer-implemented system for optimizing allocation of speech recognition tasks among multiple speech recognizers and combining recognizer results is described. An allocation determination is performed to allocate speech recognition among multiple speech recognizers using at least one of an accuracy-based allocation mechanism, a complexity-based allocation mechanism, and an availability-based allocation mechanism. The speech recognition is allocated among the speech recognizers based on the determined allocation. Recognizer results received from multiple speech recognizers in accordance with the speech recognition task allocation are combined.

Abstract: A system and method for language variation guided operator selection is disclosed. The method discloses: initiating a dialog between a contact and a call handling system; identifying a language variation spoken by the contact; determining a skill level with respect to the language variation for each operator within a set of operators; selecting an operator whose skill level in the language variation is above a predetermined value; and transferring the dialog with the contact to the operator. The system discloses means and embodiments for implementing the method.

Abstract: A system and method for managing telephone calls is disclosed. The method discloses: calling a contact; presenting the contact with a predetermined out-calling dialog; translating the contact's vocal responses to the dialog into textual words using selected interactive voice response algorithms; connecting the contact to a human operator after a predetermined portion of the out-calling dialog with the contact is completed; and providing the operator with the textual words. In one embodiment, the system discloses all means for implementing the method. In another embodiment, the system discloses: a contact database for storing information on the contact; a dialog database containing a predetermined out-calling dialog; a call manager for calling the contact and presenting the contact with the dialog; and an interactive voice response module for translating the contact's vocal responses to the dialog into textual words and storing the words in the contact database which are accessible to the operator.

Abstract: In at least some embodiments, a system may comprise a user voice interface, a processor coupled to the user voice interface, wherein the processor interprets words spoken by a user, and a memory coupled to the processor, wherein the memory stores an email application, wherein the email application summarizes email messages and navigates a plurality of email messages according to the words spoken by the user.

Abstract: A system comprises a computer system comprising a central processing unit coupled to a memory and resource management application. A plurality of different automatic speech recognition (ASR) engines is coupled to the computer system. The computer system is adapted to select ASR engines to analyze a speech utterance based on resources available on the system.

Abstract: A system and method for call center dialog management is disclosed. The method discloses: presenting a contact with a first call center dialog segment having a current call center dialog property; receiving from the contact a contact dialog segment; identifying a dialog property keyword within the contact dialog segment; replacing the current call center dialog property with a new call center dialog property in response to the dialog property keyword; and presenting a second call center dialog segment having the new call center dialog property to the contact. The system of the present invention, discloses means for implementing the method.

Abstract: A method for finding text reading order in a document such as a scanned newspaper or magazine includes the steps of pruning unnecessary text zones using semantic analysis (40), using text correlation measures to cluster zones (41), and then finding a reading order within each of the clusters (42).

Abstract: An article is extracted from a document using a decision combiner to process a plurality of reading order alternatives. The text flow analysis generates the plurality of reading order alternatives of separate body text regions.

Abstract: A method, apparatus, and system are disclosed for computer assisted document modification. One embodiment is a method for software execution. The method automatically extracts articles, in a first phase, from documents to generate different zones of the articles. Different zones of the extracted articles are displayed. In a second phase, plural different zones are manually modified with a document correction tool.

Abstract: A system and method for quality of service management within a call handling system is disclosed. The method discloses: initiating a dialog between a contact and an Interactive Voice Response (IVR) module; matching the contact with a predefined contact category; retrieving a predefined quality of service level associated with the contact category; and processing the dialog in accordance with the quality of service level. The system discloses means for effecting the method.

Abstract: A system and method for extracting demographic information from a contact is disclosed. The method discloses: initiating a dialog between a contact and a call handling system; selecting a set of demographic characteristics; assigning a set of acoustic confidence scores; assigning a set of substantive confidence scores; combining the acoustic and substantive confidence scores; and tailoring information presented to the contact using the set of combined confidence scores. The system discloses: an Interactive Voice Response module for initiating a dialog between a contact and a call handling system, and selecting a set of demographic characteristics; an acoustic classifier for assigning a set of acoustic confidence scores; a substantive classifier for assigning a set of substantive confidence scores; and a data combiner for combining the acoustic and substantive confidence scores. The Interactive Voice Response module also tailors information presented to the contact using the set of combined confidence scores.

Abstract: A method of detecting advertisements within a document comprising: identifying at least one region within an electronic version of the document; determining at least one property of the at least one region; and determining whether a region is an advertisement according to rules applied to the properties of the at least one region.

Abstract: A method, apparatus, and system are disclosed for computer assisted document modification. One embodiment is a method for software execution. The method automatically extracts articles, in a first phase, from documents to generate different zones of the articles. Different zones of the extracted articles are displayed. In a second phase, plural different zones are manually modified with a document correction tool.

Abstract: A method, apparatus, and system are disclosed for computer assisted document analysis. One embodiment is a method for software execution. The method includes selecting, in response to user input, criteria in a character recognition engine to identify suspect errors in scanned documents; executing the engine on a subset of the scanned documents to determine an accuracy of error detection using the criteria; and adjusting, in response to user input, the criteria to adjust the accuracy of identifying suspect errors.

Abstract: A speech recognition system comprises exactly two automated speech recognition (ASR) engines connected to receive the same inputs. Each engine produces a recognition output, a hypothesis. The system implements one of two (or both) methods for combining the output of the two engines. In one method, a confusion matrix statistically generated for each speech recognition engine is converted into an alternatives matrix in which every column is ordered by highest-to-lowest probability. A program loop is set up in which the recognition outputs of the speech recognition engines are cross-compared with the alternatives matrices. If the output from the first ASR engine matches an alternative, its output is adopted as the final output. If the vectors provided by the alternatives matrices are exhausted without finding a match, the output from the first speech recognition engine is adopted as the final output. In a second method, the confusion matrix for each ASR engine is converted into Bayesian probability matrix.

Abstract: An automated document processing system is configured to normalize zones obtained from a document, ant to extract articles from the normalized zones. In one configuration, the system receives at least one zone from the document, and applies at least one zone-breaking factor, thereby creating normalized sub-zones within which text lines are consistent with the at least one zone-breaking factor. The normalized sub-zones may be evaluated to obtain a reading order. Adjacent sub-zones are joined if text similarity exceeds a threshold value. Weakly joined sub-zones are separated where indicated by a topic vectors analysis of the weakly joined sub-zones.

Abstract: A method, apparatus, and system are disclosed for tuning software engines. In one exemplary embodiment, a method for software execution includes activating copies of an un-tuned software engine capable of generating a solution domain to a given input; tuning a first un-tuned software engine to generate a first subset of the solution domain in response to the given input; and tuning a second un-tuned software engine to generate a second subset of the solution domain in response to the given input.