[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: sound juicer over-range



* Aubrey Raech <aubreyraech@gmail.com> [130307 04:15]:
> On Wed, 6 Mar 2013 23:14:47 +0000
> "Russell L. Harris" <rlharris@broadcaster.org> wrote:
... 

>> It occurs to me that I could use an editor such as Audacity to
>> reduce the signal amplitude; but I do not know whether that
>> approach would give results which are sonically-pleasing.
...
> if you do want to reduce the amplitude of the entire audiobook, I
> think the easiest way to do this is with 'sox'. It wouldn't be hard
> to write a quick script to slightly reduce the amplitude of every
> track (ripped to .wav first perhaps with cdparanoia). In fact, when
> I rip audiobooks, I usually use sox to convert every track to mono
> and 22050hz, which isn't hard with a little bash.

I tend to be overly-cautious and not terribly prone to
experimentation; but here I suppose that the expenditure of an hour in
setup and experimentation on a track or two is in order.

Numerous authoritative sources warn of the dastardly effects of
over-range when using a digital recorder; this is one reason for a
general shift to 24-bit recorders over the past five years or so --
with a dynamic range of 24 bits, adjustment of the recording gain
becomes rather non-critical, because the dynamic range of music can be
captured accurately with less than 16 bits.

So I do not expect to discover that there is a good (much less, easy)
remedy (such as scaling the entire recording) for a recording which
has been made with the gain set too high.  Rather, I expect to find
that when a 16-bit recorder receives a peak for which 16 bits is
inadequate, the samples in the neighbourhood of the peak somehow are
not recorded properly, with the result that the sound is irreparably
"scrubbled".

RLH




Reply to: