Universal remote control allowing natural language modality for television and multimedia searches and requests

A remote control device comprising:

• a handheld case having a communication interface through which control instructions are issued to a remote component;
• a display screen disposed in said case;
• a microphone disposed in said case;
• a speech recognizer system coupled to said microphone;
• a user profile data store for storing information selected from the group consisting of prior use information, preference information and combinations thereof;
• a dialog manager coupled to said speech recognizer system, to said user profile data store, and to said display screen for issuing control commands through said communication interface and for displaying information on said display screen based at least in part on information obtained from said user profile data store;
• and
• a natural language parser in communication with said speech recognizer system, said parser being operative to extract a semantic content from said processed speech input and to access semantic representation of an electronic program guide and using said extracted semantic content to generate control instructions for said remote control, such that the natural language parser is a task-based parser employing a grammar comprising a plurality of frames having slots representing semantic structure of said electronic program guide, wherein the natural language parser further comprises a local parser adapted to identify predetermined task-related fragments in speech input received from said microphone, and a global parser adapted to receive task-related fragments and to extract overall semantics from the task-related fragments.

A remote control system for controlling at least one audio-video component comprising:

• a handheld case;
• a microphone disposed in said case for receiving speech input from a user;
• a communication system disposed in said case for transmitting data signals to a location remote from said handheld case;
• a speech recognizer for converting said speech input to text output;
• a memory for storing a semantic representation of an electronic program guide;
• a natural language parser in communication with said speech recognizer and with said memory, said parser being operative to extract semantic content from said processed speech input and to access semantic representation of said electronic program guide using said extracted semantic content to generate control instructions for said audio/video component, such that the natural language parser is a task-based parser employing a grammar comprising a plurality of frames having slots representing semantic structure of said electronic program guide, wherein the natural language parser further comprises a local parser adapted to identify predetermined task-related fragments in said speech input, and a global parser adapted to receive task-related fragments and to extract overall semantics from the task-related fragments.