To this end, a multiple microphone array with controllable gain and delay is utilized.
We propose robust distant speech recognition by combining multiple microphone-array processing with position-dependent cepstral mean normalization (CMN).
We combine the delay-and-sum beamforming with multiple-decoder processing or single-decoder processing, which is termed multiple microphone-array processing.
The multiple microphone-array processing using a single decoder needs about one-third the computational time of that using multiple decoders without degrading speech recognition performance.
The proposed multiple microphone-array processing using multiple decoders with position-dependent CMN achieved a 3.2% improvement (50% relative error reduction rate) over the delay-and-sum beamforming with conventional CMN (i.e., the conventional method).
The paper focuses on the design of a practical system pipeline for always-listening, far-field spoken command recognition in everyday smart indoor environments that consist of multiple rooms equipped with sparsely distributed microphone arrays.
We refers to the proposed approach as DNNSpatial, because it selectively combines LPS input features of reverberant speech obtained from multiple microphones in an array and map them into the expected output LPS features of anechoic reference speech.
We propose a speech dereverberation framework, namely DNNSpatial, by selectively combining input LPS features of reverberant speech from multiple microphones in an array and map them into the expected output LPS features of anechoic reference speech based on DNNs.
First, to accomplish simultaneous speech dereverberation and beamforming, we propose a framework, namely DNNSpatial, by selectively concatenating log-power spectral (LPS) input features of reverberant speech from multiple microphones in an array and map them into the expected output LPS features of anechoic reference speech based on a single deep neural network (DNN).
In this paper, we first propose a speech dereverberation framework, namely DNNSpatial, by selectively combining input LPS features of reverberant speech from multiple microphones in an array and map them into the expected output LPS features of anechoic reference speech based on DNNs.
