Talker collisions in an auditory scene

摘要: From a plurality of received voice signals, signal interval in which there is talker collision between at least first and second detected. A processor receives positive detection result processes, response to this, one the signals with aim making it perceptually distinguishable. mixer mixes supply an output signal, wherein processed signal(s) replaces corresponding signals. In example embodiments, content shifted away from frequency or time. The invention may be useful conferencing system.