GG: sound restoration

[Mary Jo Watts <mwatts@rci.rutgers.edu>]

On Sat, 18 Jul 1998, Tim Conway wrote:

>(4) What are the main problems facing someone trying to restore a sound?

Well, I've never had to restore a sound recording so I might be off in
the way that I express the actual physical problems involved but I have
pretty extensive experience working with digital archives so I'll give
this a college try.

A point that's not really been mentioned in this thread is that
restorations are always a product of the time-period in which they were
done-- tastes change and ideas about what restoration should do change
with fashion.  

Restoration and translation have a lot in common philosophically.  As
you tinker with and represent the original, you have to be mindful that
you have an understanding of what the original author's/artist's/ or
technician's intentions were but you're not them. You are altering the
artifact-- be it sound, film, photos,  documents, etc. The general rule
with most academic archival work is to err on the side of caution and
the least expense.  

A couple of film examples:  1) The restored version of Hitchcock's
_Vertigo_ includes sound-effects that were not present in the original-
Jimmy Stewart's car has a rumbling digital idle as he stalks Kim Novak
and footstep sounds have been added to the Mission scene.  The idea
behind the addition of these (IMHO hideous) effects is that Hitchcock
would have used stereo (in this way) had it been available to him.  This
is the aural equivalent of Turner's stupid colorize-the-classics
campaign of years past. It may be an interesting addition or add
suspense for a general contemporary audience, but it's not Hitchcock. 
2) The restored version of _My Fair Lady_ had several "holes" in the
print they used as the base for the restoration-- holes that existed
over the opening credit sequence and over Audrey Hepburn's desperate
face in a key scene.  The problem was solved by adding digital
information to the holes-- pixels with shading patterns copied from
undamaged spots. Just fixing AH's face for a few seconds of screen time
cost about $10,000 but the restorers decided it was worth the cost.  Who
knows what wasn't?

Each archival project has its own purpose and philosophy. Here are some
basic questions I ask myself when working with documents and photos:

What kind of archival project am I creating?  Is my goal to digitally
represent the artifact in question so that it can replace the physical
use of the fragile original? Has the original been damaged and to what
extent? Am I attempting to go a step further and represent the material
digitally so that I can analyze it in a way that hadn't been previously
possible with the original? What am I willing to sacrifice in terms of
the original information to have clearer access to parts of it or to
information lost due to damage? How will I document what I have done?
How much time and money do I have?  

Most of us would like to zap that hissing and popping (BECAUSE we've
grown accustomed to digital sound) but what are we willing to give up in
the process and how will we know what's been edited for us? What would
we lose if we could edited out GG's humming? 

Are GG's Liszt transcription recordings a sort of restoration?!

-Mary Jo Watts
mwatts@rci.rutgers.edu
listowner, f_minor