Abstract / truncated to 115 words (read the full abstract)

A major goal in distant-speech recognition is to transform speech signals of a target speaker into symbols in order to trigger a dialog manager. Spatio-temporal filters, so called beamformers, usually enhance the target speaker's speech signals in a noisy and reverberant environment. However, a beamformer requires information on the target speaker's position. A source localizer provides this information, which facilitates steering a beam into the direction of the target speaker. Unfortunately, the beamformer also captures noise and reverberation, especially from the target speaker's direction. To additionally reduce these artifacts, one can employ bandpass filters in order to emphasize the target speaker's harmonic components. But these bandpass filters require information on the target speaker's fundamental frequency. ... toggle 26 keywords

chirp z-transform – data association – direction of arrival – fundamental frequency – glottogram – GM-PHD – GM-CPHD – gm-cbmember – joint estimation – microphone array – multiple-target tracking – optimal subpattern assignment – pitch analysis – pitch estimation – pitch-period doubling – position-pitch algorithm – POPI – probability hypothesis density filter – relative phase-delay masking – RPDM – source localization – sparse joint parameter space – speaker separation – speaker tracking – variable-scale sampling – VSS

Follow @eurasip