Abstract: A method and system for training a user authentication by voice signal are described. In one embodiment, a set of feature vectors are decomposed into speaker-specific recognition units. The speaker-specific recognition units are used to compute distribution values to train the voice signal. In addition, spectral feature vectors are decomposed into speaker-specific characteristic units which are compared to the speaker-specific distribution values. If the speaker-specific characteristic units are within a threshold limit of the speaker-specific distribution values, the speech signal is authenticated.

Abstract: A method and system for training a user authentication by voice signal are described. In one embodiment, a set of feature vectors are decomposed into speaker specific recognition units. The speaker-specific recognition units are used to compute distribution values to train the voice signal. In addition, spectral feature vectors are decomposed into speaker-specific characteristic units which are compared to the speaker-specific distribution values. If the speaker-specific characteristic units are within a threshold limit of the speaker-specific distribution values, the speech signal is authenticated.

Abstract: A method and system for training a user authentication by voice signal are described. In one embodiment, a set of feature vectors are decomposed into speaker-specific recognition units. The speaker-specific recognition units are used to compute distribution values to train the voice signal. In addition, spectral feature vectors are decomposed into speaker-specific characteristic units which are compared to the speaker-specific distribution values. If the speaker-specific characteristic units are within a threshold limit of the speaker-specific distribution values, the speech signal is authenticated.

Abstract: A method and system for training a user authentication by voice signal are described. In one embodiment, a set of feature vectors are decomposed into speaker-specific recognition units. The speaker-specific recognition units are used to compute distribution values to train the voice signal. In addition, spectral feature vectors are decomposed into speaker-specific characteristic units which are compared to the speaker-specific distribution values. If the speaker-specific characteristic units are within a threshold limit of the speaker-specific distribution values, the speech signal is authenticated.

Abstract: Techniques for providing semantic search of a data store are disclosed. A similarity metric of a document comprising the data store to a concept represented in a semantic model derived at least in part from a reference source that includes content not included in the data store is determined. A relevance metric of a search query to the concept is computed. The similarity metric and the relevance metric are used to determine, at least in part, a ranking of the document with respect to the search query.

Abstract: Methods for beat synchronization between media assets are described. In one embodiment, beat synchronized media mixes can be automatically created. By way of example, a beat synchronized event mix can be created by selecting a plurality of media assets, arranging the media assets into an unsynchronized media mix, determining the a profile of each of the media assets in the media mix, automatically beatmatching the beats of adjacent media assets in the media mix, and automatically beatmixing the beats of adjacent beatmatched media assets to create the beat-synchronized media mix. The media assets that can be used include both audio and video media. Media assets are selected based on a specific set of media asset selection criteria, which can include music speed or tempo, music genre, music intensity, media asset duration, user rating, and music mood.

Abstract: A method and system for training a user authentication by voice signal are described. In one embodiment, a set of feature vectors are decomposed into speaker-specific recognition units. The speaker-specific recognition units are used to compute distribution values to train the voice signal. In addition, spectral feature vectors are decomposed into speaker-specific characteristic units which are compared to the speaker-specific distribution values. If the speaker-specific characteristic units are within a threshold limit of the speaker-specific distribution values, the speech signal is authenticated.

Abstract: A dictation computer that includes a fan speed regulator is described. The fan speed regulator monitors a speech recognition unit to determine when the speech recognition unit is activated. Upon detection that the speech recognition unit is activated, the fan speed regulator ducks the speed of a cooling fan embedded within the dictation computer to an optimized speed of rotation over a delay time interval. The fan speed regulator may include components to adapt the optimized speed and delay time to the characteristics of the dictation computer and the user. Other embodiments are also described.

Abstract: Techniques for providing semantic search of a data store are disclosed. A similarity metric of a document comprising the data store to a concept represented in a semantic model derived at least in part from a reference source that includes content not included in the data store is determined. A relevance metric of a search query to the concept is computed. The similarity metric and the relevance metric are used to determine, at least in part, a ranking of the document with respect to the search query.

Abstract: Methods for beat synchronization between media assets are described. In one embodiment, beat synchronized media mixes can be automatically created. By way of example, a beat synchronized event mix can be created by selecting a plurality of media assets, arranging the media assets into an unsynchronized media mix, determining the a profile of each of the media assets in the media mix, automatically beatmatching the beats of adjacent media assets in the media mix, and automatically beatmixing the beats of adjacent beatmatched media assets to create the beat-synchronized media mix. The media assets that can be used include both audio and video media. Media assets are selected based on a specific set of media asset selection criteria, which can include music speed or tempo, music genre, music intensity, media asset duration, user rating, and music mood.

Abstract: Methods for beat synchronization between media assets are described. In one embodiment, beat synchronized media mixes can be automatically created. By way of example, a beat synchronized event mix can be created by selecting a plurality of media assets, arranging the media assets into an unsynchronized media mix, determining the a profile of each of the media assets in the media mix, automatically beatmatching the beats of adjacent media assets in the media mix, and automatically beatmixing the beats of adjacent beatmatched media assets to create the beat-synchronized media mix. The media assets that can be used include both audio and video media. Media assets are selected based on a specific set of media asset selection criteria, which can include music speed or tempo, music genre, music intensity, media asset duration, user rating, and music mood. A beat synchronized event mix can be subdivided into one or more event mix segments. Each event mix segment can have its own selection criteria.

Abstract: Apparatus, methods and systems for creating mosaics are described. A method is provided that includes identifying an image for presentation as a mosaic; dividing the identified image into a plurality of tiles; scoring each tile; identifying a matching image from an image dataset using the score for each tile; and rendering the mosaic using the matching images.

Abstract: A method, apparatus, and system, for scanning a first portion of a data to generate a second portion of data is provided. A control parameter relating to a level of detail associated with filtering a first portion of data is received. The filtering of the first portion of data is performed based upon the control parameter. The filtering of the first portion of data includes a rule-based filtering, a context-based filtering, a statistical-based filtering, or a semantic-based filtering. Performing the filtering provides for a reduction of a portion of the first portion of data. A second portion of data that is smaller than the first portion of data is provided based upon the filtering of the first portion of data.

Abstract: A method, apparatus, and system, for scanning a first portion of a data to generate a second portion of data is provided. A control parameter relating to a level of detail associated with filtering a first portion of data is received. The filtering of the first portion of data is performed based upon the control parameter. The filtering of the first portion of data includes a rule-based filtering, a context-based filtering, a statistical-based filtering, or a semantic-based filtering. Performing the filtering provides for a reduction of a portion of the first portion of data. A second portion of data that is smaller than the first portion of data is provided based upon the filtering of the first portion of data.

Abstract: Methods, apparatus, systems, and computer program products are provided for synthesizing speech. One method includes matching a first level of units of a received input string to audio segments from a plurality of audio segments including using properties of or between first level units to locate matching audio segments from a plurality of selections, parsing unmatched first level units into second level units, matching the second level units to audio segments using properties of or between the units to locate matching audio segments from a plurality of selections and synthesizing the input string, including combining the audio segments associated with the first and second units.

Abstract: Systems, apparatus, methods and computer program products are described for producing text-to-speech synthesis with non-speech sounds. In general, some of the pauses or silences that would otherwise be generated in synthesized speech are instead synthesized as non-speech sounds such as breaths. Non-speech sounds can be identified from pre-recorded speech that can include meta-data such as the grammatical and phrasal structure of words and sounds that precede and succeed non-speech sounds. A non-speech sound can be selected for use in synthesized speech based on the words, punctuation, grammatical and phrasal structure of text from which the speech is being synthesized, or other characteristics.

Abstract: A method and apparatus for filtering messages comprising determining a first semantic anchor corresponding to a first group of messages, for example, legitimate messages and a second semantic anchor corresponding to a second group of messages, for example, unsolicited messages. Determining a vector corresponding to an incoming message; comparing the vector corresponding to the incoming message with at least one of the first semantic anchor and the second semantic anchor to obtain a first comparison value and a second comparison value; and filtering the incoming message based on the first comparison value and the second comparison value.

Abstract: A method, apparatus, and signal-bearing medium that files data in a destination based on one or more criteria. In various embodiments, the data may be email, email attachments, faxes, telephone messages, downloaded data or programs, audio, video, scanned images, photographs, blocks of text, or other data. In an embodiment, a training mode and an automatic mode are provided. During the training mode, a user is presented with data and a recommended destination, and the user provides feedback that is used to train the criteria. During an automatic mode, the data may be transferred to the destination with or without user confirmation.

Abstract: A method and apparatus for filtering messages comprising determining a first semantic anchor corresponding to a first group of messages, for example, legitimate messages and a second semantic anchor corresponding to a second group of messages, for example, unsolicited messages. Determining a vector corresponding to an incoming message; comparing the vector corresponding to the incoming message with at least one of the first semantic anchor and the second semantic anchor to obtain a first comparison value and a second comparison value; and filtering the incoming message based on the first comparison value and the second comparison value.

Abstract: A method and apparatus is provided for generating speech that sounds more natural. Determining whether information in a current sentence is new or previously given is performed based on a semantic relationship between the current sentence and a number of preceding sentences. A word prominence for the synthetic speech to a word in the current sentence is assigned in accordance with the information determination. A speech representative of the current sentence can be generated. In one embodiment, word prominence and latent semantic analysis are used to generate more natural sounding speech. A method for generating speech that sounds more natural may comprise generating synthesized speech having certain word prominence characteristics and applying a semantically-driven word prominence assignment model to specify word prominence consistent with the way humans assign word prominence.