Stable encoding of sounds over a broad range of statistical parameters in the auditory cortex