Vocollect, Inc.

Carnegie Mellon University

Arthur E McNair

James Randall Logan

Ryan Anthony Zoschg

Roger Graham Byford

Kurt Charles Miller

Vanessa Cassandra Sanchez

Philip E Russell

Allan Strane

Bernhard Suhm

Alexander Waibel

Alex H Waibel

Provided herein are a multi-microphone system and method including a controller, a plurality of transducers each operable within a unique sensitivity range, and corresponding microphone units. The controller receives a sound signal output from a first microphone unit that corresponds to a microphone unit having a transducer with the highest sensitivity. The controller analyzes the sound signal output to identify a first parameter of the sound signal output and determines if the first parameter satisfies pre-defined criteria. In an instance in which the first parameter satisfies the pre-defined criteria, the controller outputs the sound signal output of the selected first microphone unit as the output of the multi-microphone system. Otherwise, the controller receives a sound signal output from a second microphone unit comprising a corresponding transducer with a sensitivity less than the first microphone unit but greater than remaining transducers.

Multi-microphone system and method

A portable terminal has a network interface that receives a set of instructions having a sequence of at least one location and audio properties associated with the at least one location from a server. An audio circuit receives audio signals picked up by a microphone and processes the audio signals in a manner defined by the audio properties associated with the at least one location. A speech recognition module receives processed signals from the audio circuit and carries out a speech recognition process thereupon.

Utilization of location and environment to improve recognition

A headset for communication with a device is configured for processing audio signals captured by the headset to detect user speech and transmitting, to the device, sampled representations of the captured audio signals generally only when user speech is detected.

Wireless headset for use in speech recognition environment

The method of the present invention is comprised of the steps of locating an error within a recognition hypothesis. A repair hypothesis is generated for the located error from a secondary input signal. At least a portion of the recognition hypothesis is correlated with the repair hypothesis to produce a new hypothesis for the located error. The located error is replaced with the new hypothesis. It has been determined that when the primary and secondary input signals are cross-modal, the chances that the error will be repeated are so low that the correlation step becomes optional.

Method and apparatus for correcting and repairing machine-transcribed

A method of repairing machine-recognized speech is comprised of the steps of receiving from a recognition engine a first n-best list of hypotheses and scores for each hypothesis generated in response to a primary utterance to be recognized. An error within the hypothesis having the highest score is located. Control signals are generated from the first n-best list which are input to the recognition engine to constrain the generation of a second n-best list of hypotheses, and scores for each hypothesis, in response to an event independent of the primary utterance. The scores for the hypotheses in the first n-best list are combined with the scores for the hypotheses in the second n-best list. The hypothesis having the highest combined score is selected as the replacement for the located error.

Growing community of inventors

Pittsburgh, PA, United States of America

Arthur E McNair