1

I record audio streams from internet radio stations.

The recordings are usually about an hour long ( mpeg or flac format ) and each recording contains several songs / pieces of music.

Is there software that could take the recordings as input, discern / detect the individual song boundaries ( as a best effort of course ) and output the songs / pieces contained in each input file as individual audio files?

1

Possible solutions to the above:

i) When the stream contains metadata ( SHOUTcast or Icecast formats ) there is software that can actually leverage this information and chunk up the stream into individual tracks ( example 'streamripper', 'StreamRipStar' ...) complete with artist and title info incorporated into the filenames for each track.

ii) When the recording was made from a stream without metadata, then one must identify points of silence within the stream. I.e. look for points where the signal drops below some threshold dB level for a certain period of time. 'Audacity' provides a menu to aid defining and automatically finding such points in the stream ( Audacity silence finder ).

Not the answer you're looking for? Browse other questions tagged or ask your own question.