WEIGHTED AUTOCORRELATION-BASED F0 ESTIMATION FOR DISTANT-TALKING INTERACTION WITH A DISTRIBUTED MICROPHONE NETWORK

A distant-talking scenario is addressed, where a distributed microphone network provides multi-channel input sequences to process for speaker modeling purposes. Possible related appli- cations are speaker tracking and distant-talking speech recogni- tion, given a noisy and reverberant environment with one or more speakers. The paper investigates on the use of a multi-channel version of a Weighted Autocorrelation(WAUTOC)-based F0 esti- mation method, with the purpose of deriving a common excitation model. Experiments conducted on a real database show the advan- tages and the robustness of the proposed method in extracting the fundamental frequency with no regard about the microphone and talker position as well as the head orientation.