Special issue on speech separation and recognition in multisource environments

One of the chief difficulties of building distant-microphone speech recognition systems for use in `everyday' applications is that the noise background is typically `multisource'. A speech recognition system designed to operate in a family home, for example, must contend with competing noise from televisions and radios, children playing, vacuum cleaners, and outdoors noises from open windows. Despite their complexity, such environments contain structure that can be learnt and exploited using advanced source separation, machine learning and speech recognition techniques such as those presented at the 1st International Workshop on Machine Listening in Multisource Environments (CHiME 2011).