update to processing audio post

Wanted to add an update to the audio processing post. Found that the search box in the blogger bar at the top of this blog was not working too well. Google site search - the "search hnsws.blogspot.com" option at the top of the page - was working better. I'm not removing the blogger bar just yet, because some time ago the opposite was true, with site search not returning proper results.

Anyway, here is the update. To the info in the earlier post of 2006, this auto-trim-crop tip has to be added. Also, in Sound Forge version 7 (and above?) the normalization has been tweaked a bit, so that -12 dB is almost as loud as -10 dB in the earlier Sound Forge 4.x. So, in batch processing, the processing settings look like this:



One more point to be mentioned is that when processing audio sent for regional language broadcasts, I have to select the regions with similar levels based on visual inspection and normalize them separately - m on Sound Forge drops a marker at current cursor position, and double-click on a region selects it.



Filenames for these (other language broadcasts) are also in a pattern, this can be easily learnt by looking at the older files.