Abstract: An embodiment of the present invention provides a speech recognition engine that utilizes portable voice profiles for converting recorded speech to text. Each portable voice profile includes speaker-dependent data, and is configured to be accessible to a plurality of speech recognition engines through a common interface. A voice profile manager receives the portable voice profiles from other users who have agreed to share their voice profiles. The speech recognition engine includes speaker identification logic to dynamically select a particular portable voice profile, in real-time, from a group of portable voice profiles. The speaker-dependent data included with the portable voice profile enhances the accuracy with which speech recognition engines recognize spoken words in recorded speech from a speaker associated with a portable voice profile.

Abstract: An embodiment of the present invention provides a speech recognition engine that utilizes portable voice profiles for converting recorded speech to text. Each portable voice profile includes speaker-dependent data, and is configured to be accessible to a plurality of speech recognition engines through a common interface. A voice profile manager receives the portable voice profiles from other users who have agreed to share their voice profiles. The speech recognition engine includes speaker identification logic to dynamically select a particular portable voice profile, in real-time, from a group of portable voice profiles. The speaker-dependent data included with the portable voice profile enhances the accuracy with which speech recognition engines recognize spoken words in recorded speech from a speaker associated with a portable voice profile.

Abstract: A method and system for correlating input data that includes configuration data. The configuration data describes at least one hierarchy including a first hierarchy. Each hierarchy includes at least two nodes. Each node includes a value and is designated as an ontology end point consisting of a unique positive integer. For each hierarchy, a unique relationship number RN is stored in a medium for each pair of parent-child nodes of each hierarchy. RN is a function of X and Y which respectively denote the ontology end point of the parent and child node of each pair of parent-child nodes. Responsive to a query that identifies a measure of correlation between the first hierarchy and other information, the measure of correlation is computed by retrieving and utilizing each stored relationship number of the first hierarchy. Information relating to the measure of correlation is outputted.

Abstract: Consistent with an embodiment of the invention, a communication device enables real-time communications by generating a representation of an analog input signal that is suitable for communication over a real-time communication channel to a remote communication device. In addition, the communication device stores and processes an alternative representation of the input, which is communicated to the remote device in a reliable, out-of band channel. Accordingly, the remote communication device initially stores the real-time data, but later automatically replaces the real-time data with the enhanced data that it receives over the reliable, out-of-band communication channel. The communication device similarly stores, and replaces, data received from the remote communication device.

Abstract: A real-time communication device with an asymmetric mode is disclosed. During a real-time communication session, a communication application executing on the communication device detects an upstream communication channel having insufficient bandwidth to support real-time communications in an upstream direction. Consequently, the communication application enters into an asymmetric mode. In asymmetric mode, as the communication application continues to receive data over a downstream communication channel, the communication application allows a user to record a media clip, and communicate the media clip to a remote communication device over the upstream communication channel at a transfer rate supported by the available bandwidth of the upstream communication channel.

Abstract: Methods and systems for enhancing and/or improving one's memory are disclosed. In one embodiment of the invention, a computer-implemented memory assistance system is comprised of a client and a memory assistance server. The memory assistance server includes a signal processing server to receive and process media signals (e.g., audio, video, or photographs) and associated data received from the client. The signal processing server includes a variety of servers or services for processing signals and/or data. Each signal processing server or service identifies characteristics of the received media signal that are relevant to aiding a user in recalling and/or reinforcing an existing memory of a particular event or fact. In addition, the memory assistance server includes presentation logic to generate one or more media clips from the received media signal.