How do I find a dropout or DC segment? -Sound Forge - PC

Subject:How do I find a dropout or DC segment? Posted by: dgeston Date:3/18/2005 9:58:32 AM I am looking for a way in Sound Forge to automatically locate a dropout, or more generically a user-determined number of sequential samples that are all the same. I have tried the Find Silence Between Phrases and it seems to look for periods of silence that are longer than the dropouts I am experiencing. I've also not found anything in the forums that specifically address this issue. The problem arises when I am copying digitally from a DAT or CD master, and there are digital errors on the tape or scratches on the CD that the error correction algorithms can't handle. The result is either bad sound or a complete (but short) dropout that lasts for varying amounts of time generally not longer than several hundred milliseconds. In either case there are segments created that are all zeros (silence) or all the same value (DC). It's sort of an inverse problem to clipped-peak detection. Is there any way to automatically find such beasties? Thanks in advance for any help anyone can offer. -Doug BTW: I just upgraded to SF 8.0, and the tests I did were with that version Message last edited on3/19/2005 10:02:21 AM bydgeston.

I am looking for a way in Sound Forge to automatically locate a dropout, or more generically a user-determined number of sequential samples that are all the same. I have tried the Find Silence Between Phrases and it seems to look for periods of silence that are longer than the dropouts I am experiencing. I've also not found anything in the forums that specifically address this issue.

The problem arises when I am copying digitally from a DAT or CD master, and there are digital errors on the tape or scratches on the CD that the error correction algorithms can't handle. The result is either bad sound or a complete (but short) dropout that lasts for varying amounts of time generally not longer than several hundred milliseconds. In either case there are segments created that are all zeros (silence) or all the same value (DC). It's sort of an inverse problem to clipped-peak detection.

Is there any way to automatically find such beasties? Thanks in advance for any help anyone can offer.

-Doug

BTW: I just upgraded to SF 8.0, and the tests I did were with that version