Abstract: A digitally-sound-encoded filmstrip has digital data recorded along the length of a soundtrack, said data being representative of a particular data format. The data format has a plurality of data fields which are filled with data representing sound values and other digital information. Each data field has redundant data to enable the detection and correction of data in the data field upon decoding of the formatted data. The data format also features a plurality of data blocks wherein each data block is made up of a group of data fields wherein at least one data field contains data redundant to the entire data block to enable the detection of errors made in correction of errors in the individual data fields. Successive sound samples are recorded in mutually remote locations along the length of the soundtrack.