This is a methodology for the future. Carl Haber and his crew at the
Lawrence Berkely National Lab developed the system that Patrick uses. So
does the Library of Congress. it's a great way to overcome many of the
limitations on problem recordings of various types.
It's still in development and not available to the general public yet, as
far as I know. It is also excruciatingly expensive.
Needle in groove playback still sounds better. I assume they are also
working on improving their digital modeling of the physics of playback.
As Mort Sahl said, "The future lied ahead." One reason it is so important
to preserve off-the (groove?) wall recording media in a protective, stable
environment is to allow time for technological developments to catch up.
Steve Smolian
-----Original Message-----
From: Tom Fine
Sent: Thursday, June 21, 2012 7:41 AM
To: [log in to unmask]
Subject: Re: [ARSCLIST] audio from pictures
Hi Michael:
What you say makes sense about the sound of "paper playing." Did you notice
it's a pretty steady
pattern, sounds to me like once a revolution. So my question is, does any
modern software "cancel
out" a regular/steady but complex background noise like that? If so, would
it improve audibility?
I wish Patrick's blog-site had linked to hi-rez WAV so we could all
experiment at home!
One of the first commercial uses I can envision for this is apply it to all
the super-rare but
beat-to-hell Paramount blues sides and see if better/clearer results can be
obtained than out in
print so far. It would also seem like Patrick's method works for cracked
records and cylinders
because he's basically "pasting together" the groove in a straight line, so
it should be
time-consuming but possible to line up both sides of a crack and just
pencil-over or otherwise
smooth out the crack itself.
-- Tom Fine
----- Original Message -----
From: "Michael Shoshani" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Wednesday, June 20, 2012 10:50 PM
Subject: Re: [ARSCLIST] audio from pictures
> On 06/20/2012 06:13 PM, Tom Fine wrote:
>>
>> Question -- is there an algorhythm to remove the modulation noise behind
>> the voice? It seems to be repeated modulation, just like if a disk were
>> played and the groove was "listing" to one side, sort of a
>> woosh-woosh-woosh. Because it's a relatively uniform modulation, I'm
>> wondering if it can be "zero'd out" to improve voice audibility?
>
> This actually sounds as though it could be the grain of the paper on which
> the magazine was printed. It's an extremely regular, constant sound and
> one that could easily be imagined as coming from a paper disc being
> "played" by an acoustic pickup.
>
> I used to do this with labels of junky records on my Victrola XVI when I
> was a teenager, cutting spirals into the labels, which is how I recognized
> the pattern.
>
> I don't know if Dr. Feaster is on or reads ARSCLIST. I know he used to be
> a fairly heavy participant in 78-L years ago.
>
> Michael Shoshani
> Chicago
>
|