Abstract: Text strings from computer-based applications are converted from multiple external formats into an internal format. The text strings are stored in a terminology database. Translations for the test strings are also stored in the terminology database and associated with the corresponding text strings. In one aspect, text strings are converted into a export format for translation into one or more target languages. The translations are converted into the appropriate external formats and output to the applications that supplied the text strings. In another aspect, the terminology database is queried for an appropriate translation for each newly converted text string. In still another aspect, metadata for the text strings is created and associated with the text strings in the terminology database, and used as query criteria to determine the appropriate translation. The metadata may be exported in conjunction with the text strings.

Abstract: A method and system are provided for processing an extrapolated signal including a number of consecutive replacement frames. The method comprises attenuating a portion of the extrapolated signal when the extrapolated signal reaches a predetermined duration. The attenuating produces an output signal having an attenuated portion, wherein the output signal includes the number of consecutive replacement frames. Each of the consecutive frames within the attenuated portion is attenuated by applying an attenuation window with a starting magnitude value of approximately 1 and including a unique ending magnitude. The unique ending magnitudes decrease over time.

Abstract: A system that includes a speech synthesizer and a computer system that includes a basic input output system (BIOS). The BIOS is configured to cause the computer system to translate pre-boot display information, in response to detecting the speech synthesizer, so as to provide one or more serial data signals associated with the information to the speech synthesizer. The speech synthesizer is configured to generate one or more audible outputs associated with the information in response to receiving the one or more signals.

Abstract: An arrangement for adapting statistical parsers to new data using a mathematical transform, particularly a Markov transform. In particular, it is assumed that an initial statistical parser is available and a batch of new data is given. The initial model is mapped to a new model by a Markov matrix, each of whose rows sums to one. In the unsupervised setup, where “true” parses are missing, the transform matrix is obtained by maximizing the log likelihood of the parses of test data decoded using the model before adaptation. The proposed algorithm can be applied to supervised adaptation, as well.

Abstract: A customer communication is responded to by receiving an utterance from the customer at an agent that executes on a data processing system. The agent uses a knowledge base that includes information extracted from one or more exemplary conversations to generate a response to the received utterance. The agent then sends the generated response to the customer.

Abstract: A process for generating with unification based grammars such as Lexical Functional Grammars which uses construction and analysis of generation guides to determine internal facts and eliminate incomplete edges prior to constructing a generation chart. The generation guide can then be used in the construction of the generation chart to efficiently generate with unification-based grammars such as Lexical Functional Grammars. The generation guide is an instance of a grammar that has been specialized to the input and only contains those parts of the grammar that are relevant to the input. When the generation guide is analyzed to determine internal facts a smaller generation chart is produced.

Abstract: Embodiments of the invention achieve increased compression of audio data in comparison to prior art ADPCM compression schemes using modest processing power and resources. For one embodiment, an asymmetric ADPCM encoding scheme is implemented to increase apparent encoding resolution for a specified number of encoding bits. Additionally, or alternatively, such techniques as pattern recognition and encoding, as well as calculation simplification, are employed to increase data compression of audio data.

Abstract: A system and method for voice quality analysis include the ability to receive packets in a voice stream and to generate a receipt indicator for the packets. The system and method also include the ability to substitute a reference voice sample for the voice data in the packets and to compare the voice data in the voice-substituted packets to the reference voice sample to determine voice quality.

Abstract: Legality expressions are systematically pre-processed, organized, and stored to achieve faster real-time response, improved predictability, and increased reliability for queries against a large volume of legality expressions. Exponential improvements in both the time to locate the set of legality expressions matching specified search criteria and the processing costs of evaluating the request against the matching legality expressions are achieved using the disclosed systems, devices, and methods. The systems, devices, and methods are unique to the optimization of legality expression processing, but they can also enable the use of other optimization techniques for processing large amounts of data.

Abstract: The present invention is a system and method for normalizing a discourse representation structure (DRS). The elements of the structure are rewritten and sorted in a way such that structures which may appear different but are nonetheless equivalent can be associated with the same, normalized representation. The present invention can also include a data structure for a DRS. The DRS is represented by an array of boxes, each having a set of elements which in turn has a predefined structure suitable for representing a wide variety of linguistic information.

Abstract: Multimodal utterances contain a number of different modes. These modes can include speech, gestures, and pen, haptic, and gaze inputs, and the like. This invention use recognition results from one or more of these modes to provide compensation to the recognition process of one or more other ones of these modes. In various exemplary embodiments, a multimodal recognition system inputs one or more recognition lattices from one or more of these modes, and generates one or more models to be used by one or more mode recognizers to recognize the one or more other modes. In one exemplary embodiment, a gesture recognizer inputs a gesture input and outputs a gesture recognition lattice to a multimodal parser. The multimodal parser generates a language model and outputs it to an automatic speech recognition system, which uses the received language model to recognize the speech input that corresponds to the recognized gesture input.

Abstract: A list presentation method. The list presentation method can include the steps of: dynamically grouping selected items in a list based on sequentially positioned symbols in the items which are common to one another; labeling each group of selected items; audibly presenting each group label through a speech user interface; and, responsive to a selection of one of the presented group labels, presenting through the speech user interface items in a group corresponding to the selected group label.

Abstract: A method for verifying a speaker by a computer is specified in which the speaker utters at least one code word consisting of a random combination of syllable units previously supplied by the speaker. A comparison of this uttered code word with at least one pattern verifies the speaker if the at least one pattern matches the uttered code word with a predetermined minimum quality. The code word is designed in such a manner that it has at least two syllable units, at least one syllable unit being a component of a further code word.

Abstract: Components of an application that may be requested by a user in the future are identified and adapted for operation with a device of the user in response to a request for a component of the application by the user. Optionally, only the requested and identified components are adapted for operation with the requesting device, which may enable component adaptation within a specified maximum time period. A reachability analysis may be performed to identify components reachable from the requested component and components that are within a specified distance of the requested component may be identified or selected. The identified components may further be selected or identified based on historical request patterns relating to the application.

Abstract: A method of smoothing fundamental frequency discontinuities at boundaries of concatenated speech segments includes determining, for each speech segment, a beginning fundamental frequency value and an ending fundamental frequency value. The method further includes adjusting the fundamental frequency contour of each of the speech segments according to a linear function calculated for each particular speech segment, and dependent on the beginning and ending fundamental frequency values of the corresponding speech segment. The method calculates the linear function for each speech segment according to a coupled spring model with three springs for each segment. A first spring constant, associated with the first spring and the second spring, is proportional to a duration of voicing in the associated speech segment. A second spring constant, associated with the third spring, models a non-linear restoring force that resists a change in slope of the segment fundamental frequency contour.

Abstract: A method for determining if one audio signal is derived from another audio signal or if two audio signals are derived from the same audio signal compares reduced-information characterizations of said audio signals, wherein said characterizations are based on auditory scene analysis. The comparison removes from the characterisations or minimizes in the characterisations the effect of temporal shift or delay on the audio signals (5-1), calculates a measure of similarity (5-2), and compares the measure of similarity against a threshold. In one alternative, the effect of temporal shift or delay is removed or minimized by cross-correlating the two characterizations. In another alternative, the effect of temporal shift or delay is removed or minimized by transforming the characterizations into a domain that is independent of temporal delay effects, such as the frequency domain. In both cases, a measure of similarity is calculated by calculating a coefficient of correlation.

Abstract: A computer system for implementing a method for managing translations of text phrases is disclosed. The system stores text phrases in a source language as individual entries within a database. The system also stores any imported text phrases in a target language as individual entries within the database. For each text phrase in the source language stored within the database, the system is operable to communicate a notification of an exclusion of a corresponding text phrase in the target language within the database. Also for each stored text phrase stored in the source language within the database, the system is operable to communicate a notification of a modification of a text phrase in the source language as stored within the database after a storage of a corresponding text phrase in the target language within the database.

Abstract: A computer-loadable data structure is provided that represents a state-and-transition-based description of a speech grammar. The data structure includes first and second transition entries that both represent transitions from a first state. The second transition entry is contiguous with the first transition entry in the data structure and includes a last-transition value. The last-transition value indicating that the second transition is the last transition from the first state in the data structure. A method is also provided for retrieving information from a binary grammar. The method includes receiving an index into a set of transition entries and converting the index into a memory offset relative to the beginning of the binary grammar, where the offset is based on a memory offset to the beginning of the set of transition entries, the fixed size of each transition entry and the index.

Abstract: An audio encoding method of coding audio signals into a layered data stream having a first layer and a second layer is presented. The second layer serves as an enhancement of the first layer. The method involves forming an original digital audio signal, encoding the original signal to obtain a first layer signal, producing a residual signal to reflect a difference between the original signal and the first layer signal, selecting either the original signal or the residual signal for encoding, and producing a second layer signal by encoding the selected signal. Moreover, the residual signal is evaluated and, in case the result of the evaluation meets given criteria, a predetermined low-entropy signal is selected to be encoded as the second layer signal, instead of the original signal or the residual signal.

Abstract: In a method of assisting a subject to generate speech, at least one first neural impulse is sensed from a first preselected location in the subject's brain. A first preselected sound is associated with the first neural impulse. The first preselected sound is generated in an audible format. In an apparatus for assisting the subject to generate speech, at least one sensor senses a neural impulse in the subject's brain and generates a signal representative thereof. An electronic speech generator generates a phoneme in response to the generation of the signal. An audio system generates audible sounds corresponding to the phoneme based upon the signal received from the speech generator.