Resources Contact Us Home
Speech separation with microphone arrays

Image Number 12 for United States Patent #8144896.

A system that facilitates blind source separation in a distributed microphone meeting environment for improved teleconferencing. Input sensor (e.g., microphone) signals are transformed to the frequency-domain and independent component analysis is applied to compute estimates of frequency-domain processing matrices. Modified permutations of the processing matrices are obtained based upon a maximum magnitude based de-permutation scheme. Estimates of the plurality of source signals are provided based upon the modified frequency-domain processing matrices and input sensor signals. Optionally, segments during which the set of active sources is a subset of the set of all sources can be exploited to compute more accurate estimates of frequency-domain mixing matrices. Source activity detection can be applied to determine which speaker(s), if any, are active. Thereafter, a least squares post-processing of the frequency-domain independent components analysis outputs can be employed to adjust the estimates of the source signals based on source inactivity.

  Recently Added Patents
Generating a representation of an object of interest
High conductive water-based silver ink
Secure provisioning of a portable device using a representation of a key
Sparse network almanac
Methods for measuring media performance associated with adjacent track interference
Method and apparatus for editing a program on an optical disc
Optical angular position detection apparatus and method
  Randomly Featured Patents
Holder changer
Method of forming electrodes of an electronic component of chip type for connecting to the external
Heat trap
Disposable arthroscope
Hand-held massager
Pneumatic injection of powder or granule through submerged tuyeres
Water aerator
Combined hard/soft relay forwarding for hybrid-automatic repeat request (ARQ) exploitation
Magnetic pistons engine
System for mounting an object on a moving vehicle