Real and imaginary components of the discrete Fourier transform (DFT) of the residual echo in an acoustic echo cancellation system are separately estimated using mapping functions derived without an assumption of statistical independence between separate components of a near-end microphone input signal. The residual echo estimator calculates the real and imaginary components using an approximation of the microphone input signal without the echo. A different approximation is used depending on whether the near-end talker and/or the far-end talker are talking.
Low-Power Noise Characterization Over A Distributed Speech Recognition Channel
Michael E Deisher - Hillsboro OR, US Robert W Morris - Atlanta GA, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
G10L 21/02
US Classification:
704228
Abstract:
A distributed speech recognition system includes a noise floor estimator to provide a noise floor estimate to a feature extractor which provides a parametric representation of the noise floor estimate. An encoder is included to to generate an encoded parametric representation of the noise floor estimate. A front-end controller is also included to determine when at least one of the noise floor estimator, the feature extractor, and the encoder is to be turned on or off and to determine when the noise floor estimator is to provide the noise floor estimate to the feature extractor. Additionally, a decoder is included to generate a decoded parametric representation of the noise floor estimate. A noise model generator creates a statistical model of noise feature vectors based on the decoded parametric representation of the noise floor estimate.
Speech Recognition Command Via An Intermediate Mobile Device
Michael E. Deisher - Hillsboro OR, US Rajesh P. Banginwar - Hillsboro OR, US Robert C. Knauerhase - Portland OR, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
G10L 15/22
US Classification:
704275
Abstract:
According to an embodiment of the invention, a physical location of a mobile device is determined, and a determination is made that a subject device is available for command via the mobile device based at least in part on the physical location of the mobile device. Information regarding voice recognition capability of the subject device is transferred to the mobile device. A voice command is received by the mobile device, the voice command is interpreted, and an instruction is provided to the subject device based at least in part on the voice command.
Michael E. Deisher - Hillsboro OR, US Robert C. Knauerhase - Portland OR, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
H04Q 7/20 G10L 15/00
US Classification:
4554561, 704233, 381104
Abstract:
Speech recognition in mobile processor-based devices may be improved by using location information. Location information may be derived from on-board hardware or from information provided remotely. The location information may assist in a variety of ways in improving speech recognition. For example, the ability to adapt to the local ambient conditions, including reverberation and noise characteristics, may be enhanced by location information. In some embodiments, pre-developed models or context information may be provided from a remote server for given locations.
A processor-based system obtaining information about an event from schedule data, and using the information to assist speech recognition of speech occurring during at least a portion of the event.
Chaitanya Sreerama - Hillsboro OR, US Michael E. Deisher - Hillsboro OR, US Keith R. Tinsley - Beaverton OR, US Harry G. Skinner - Portland OR, US Stephen H. Hall - Hillsboro OR, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
H04B 1/00 H04B 1/04
US Classification:
455 631, 4551142, 455296
Abstract:
A method according to one embodiment identifying the frequency range of at least one active channel of at least one wireless communication RF band; identifying the frequency range of at least one clock harmonic; identifying an overlap, in whole or in part, between the frequency range of the at least one active channel and the frequency range of the at least one clock harmonic; and shifting a fundamental frequency of the at least one clock to shift the frequency range of the at least one clock harmonic out of, at least in part, the frequency range of the at least one active channel. Of course, many alternatives, variations, and modifications are possible without departing from this embodiment.
Michael E. Deisher - Hillsboro OR, US Keith R. Tinsley - Beaverton OR, US Chaitanya Sreerama - Hillsboro OR, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
H04B 1/10
US Classification:
375346, 455296
Abstract:
A method according to one embodiment for mitigating radio frequency interference by identifying system clocks, identifying active radio channels, measuring clock harmonics in or near the active radio channels, determining potential interference occurring if the clocks were moved to new fundamental frequencies, and shifting clock fundamental frequencies to reduce interference to the active radio channels based on existing interference and the potential interference of a plurality of new fundamental frequencies. Of course, many alternatives, variations, and modifications are possible without departing from this embodiment.
Adaptive Control Of Clock Spread To Mitigate Radio Frequency Interference
Harry Skinner - Portland OR, US Michael E. Deisher - Hillsboro OR, US Chaitanya Sreerama - Hillsboro OR, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
H03D 1/04
US Classification:
375346, 375371, 375376, 375373, 375294, 375215
Abstract:
In some embodiments an adaptive clocking controller determines a clock spread of a system clock that would result in a lowest total interference between a channel received by a radio receiver and the system clock. A clock generator modifies a spread of the system clock in response to the determined clock spread. Other embodiments are described and claimed.