A device, system and method for audio transmission quality assessment that occurs during the transmission. A transmission channel such as the internet is used to transmit speech that is spoken by a human speaker, captured at a first end, and transmitted over the transmission channel for reproduction at a second end. The processors at each end of the transmission channel are configured to determine one or more characteristics of the speech such as phonemes. The phonemes are transmitted over a backchannel of the transmission channel to a processor that compares the speech characteristics that were determined at both ends of the call. The participants are notified of a transmission problem that has had an effect on the intelligibility of the speech that was reproduced at the far end if the comparison does not meet a predetermined quality metric.