This review article summarizes the main difficulties encountered in Automatic Speech Recognition (ASR) when the type of communication channel is not known. This problem is crucial for the development of successful applications in promising domains such as computer telephony and cars. The main technical problems encountered are due to the speaker and the task (e.g. speaking style, Lombard reflex, vocal tract geometry), the use of microphones with different characteristics, the variable quality of the support channels (e.g. telephone channels are noisy and have different characteristics), reverberation and echoes, the variable distance and direction to the microphone introduced by hands-free recognition, and the ambient noise which distorts the input speech signals. This overview characterizes and emphasizes these problems and highlights some promising directions for future research. Finally, it presents an attempt to characterize the sensitivity of a phoneme recognizer as a function of the source of channel distortion, using the TIMIT database and several of its variants (NTIMIT, CTIMIT, FFMTIMIT).