Motorola Corporation

Motorola Mobility LLC

David John Pearce

Robert A Zurek

Joel A Clark

Jonathan A Gibbs

William M Kushner

Holly Francois

Andrew John Aftelak

James Alexander Rex

Jeffrey Arthur Meunier

Scott K Isabelle

Jeffrey M Axelrod

Dominic Chan

A system, method, and apparatus for separating speech signal from a noisy acoustic environment. The separation process may include directional filtering, blind source separation, and dual input spectral subtraction noise suppressor. The input channels may include two omnidirectional microphones whose output is processed using phase delay filtering to form speech and noise beamforms. Further, the beamforms may be frequency corrected. The omnidirectional microphones generate one channel that is substantially only noise, and another channel that is a combination of noise and speech. A blind source separation algorithm augments the directional separation through statistical techniques. The noise signal and speech signal are then used to set process characteristics at a dual input noise spectral subtraction suppressor (DINS) to efficiently reduce or eliminate the noise component. In this way, the noise is effectively removed from the combination signal to generate a good qualify speech signal.

Robust two microphone noise suppression system

A method and apparatus for smooth transitions of background noise as a mobile communication unit roams from a first communication system having a first speech encoder ( ) to a second communication system having a second speech encoder ( ). As the mobile communication unit roams into the second communication system, the first communication system will transfer the communication signal to the second communication system. Background noise from the first communication system is sampled ( ), and as the communication signal is transferred, the noise signal sampled in the first communication system is mixed in a mixer of the second communication system with background noise sampled ( ) from the second communication system. As the change in noise progresses, the noise from the first communication system will gradually be decreased and faded out while the noise from the second communication system will be gradually increased to a high power level.

Background noise contrast reduction for handovers involving a change of speech codec

Quantization unit ( ) comprises evaluator ( ) and comparator ( ) in signal processing for identifying an utterance in system ( ). The evaluator ( ) weights a first intermediate result of an operation on a first set of a plurality of speech parameters ( ) differently than a second intermediate result of an operation on a second set of the plurality of speech parameters ( ) in a weighted representation of the plurality of speech parameters ( ). The comparator ( ) employs the weighted representation of the plurality of speech parameters ( ) to determine a vector index to represent the plurality of speech parameters ( ). The quantization unit ( ), in one example, can employ split vector quantization in conjunction with the weighted representation to determine a vector index to represent the plurality of speech parameters ( ).

Speech recognition using unequally-weighted subvector error measures for determining a codebook vector index to represent plural speech parameters

Growing community of inventors

Basingstoke, United Kingdom

David John Pearce