The work described in this paper is based on the development of an experimental method of recording and reproducing audio signals. The speech signal was cut intermittently and a delayed signal was introduced in the deleted portions. The average comprehensibility was ascertained and found to be 95% and this level was maintained at all frequencies. In this way, more information may be recorded and reproduced than the ordinal)/ methods of recording and reproducing and also tape wastage may be eliminated.