Believe me, I am very sympathetic to your sentiments. More so than my posts might suggest. But with
some big transfer projects and some years of experience at this kind of "institutional" audio under
my belt, I have to say that cheaping out just leads to bad results and it's actually a waste of
money and time and irresponsible for an institution on a budget. Here are some things to consider,
and please take this as one man's opinions with as much salt as desired ...
1. a pile of cassette tapes is pretty stable, if stored properly. Better to amass some expertise and
funding than to rush headlong into a transfer project under a false idea that there's a timebomb
ticking. Yes, cassettes are not "forever" stable, but no medium is and I would suggest that most
large piles of cassettes are anywhere from 35 to 5 years old and not in immediate risk. Richard may
say different but I've had few cassettes through here that had the tape falling apart in them. Sure,
stuff stored improperly warps and has dropouts but we're not talking about oxide falling off like
can happen with old cellulose and acetate reels. This is of course not the case with all cassettes
and there are some problem types. But, in general, I'd say take your time getting funding in place
to do the job right.
2. as I've said numerous time, it is very much worth looking into outsourcing the A-D work. I've
found that for institutional clients, there are usually a few folks on-hand who are very clear on
what they want to do with the audio once it's in digital form. And, I've found these folks to be
generally highly educated and smart and thus it's not a problem for someone on staff to be come
expert at a sound-editing program. So the way we generally work to keep the client in maximum
control and the total budget within a good sweet spot is they get either the raw transfer files or,
if cleanup work is outsourced, the "final" high-rez and the raw high-rez. Then they handle the
slicing and dicing, burning to portable media, crunching to online formats, entering their own
metadata, etc. This keeps my studio time down to mainly transfer work and thus saves them a lot of
money since they have facilities already in place to do the other work -- and the key work of final
format and access files and metadata/database are handled in-house exactly to their preferred
specifications. Just to be clear, I've done projects where all this work is outsourced, but the
costs are much higher.
3. what has become clearer to me over time is that something like a field recording of an oral
history (ie a non-stellar recording) NEEDS to be transferred and handled in a very high-quality
audio chain. Things like voice-clarity and dynamics control can be very much improved for the final
product, making reference-listening much more productive and audibility much better. So apply a
barebones/dirt-cheap chain to bad audio will make things worse, whereas applying a very good chain
to very bad audio can help in many cases. I think this is one of Richard's main points.
So, while I am very sympathetic to cash-starved archives and institutions, I think it's better to
make strategic decisions about realistic funding levels and realistic expectations about what can be
digitized within those limits. Then prioritize and maybe spread out the time-horizon to get it done
-- Tom Fine
----- Original Message -----
From: "Sam Brylawski" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, August 15, 2008 9:32 AM
Subject: [ARSCLIST] The Hope of Audacity Was--Re: [ARSCLIST] Seeking recommendations for oral
history digitization equipment (fwd)
Sorry for the mistyping on Soundforge, Tom . I didn't mean to imply
that I thought that Audacity was in a league with Soundforge and any
other professional applications. I'm not qualified to say and trust
your judgment that it's not. I was just wondering whether someone on a
tight budget who is transferring spoken word from cassettes might be
OK with Audacity. I've used it for the primitive work I'm capable of
doing and it worked. Your reference to it was so dismissive that it
led me to question whether Audacity might corrupt files in some way. I
do admit, however, that I'm sensitive to over-tooling, or whatever the
expression is. That is, buying a Ferrari to drive to the grocery
On Fri, Aug 15, 2008 at 6:22 AM, Tom Fine <[log in to unmask]> wrote:
> Hi Sam:
> No reason at all they shouldn't use Soundforge! That's why I recommend it.
> It's easy to learn, easy to use and perfect for bringing in 2 channels and
> now more channels at a time. Plus the latest version comes with very nice
> processing and restoration tools that, again, are not hard to learn or use.
> And as I alluded to, the latest version (9) finally allows for more than 2
> channels at a time -- I currently have it set up for 4 channels at a time.
> An academic or institutional customer should be able to find the
> academic-discount price for any software, which is cheap compared to the
> cost of redoing a project due to bad work.
> My experience with Audacity was that it's akin to the graphic user
> interfaces floating around for Linux -- it's klunky and primative and you're
> fooling yourself if you think it's in the league of a real-deal,
> paid-professional program. It's freeware, this is something that people with
> no budget use because they have to, which is fine but non-ideal for doing
> efficient and high-quality transfer work. If I were a guy wanting to get my
> semi-worn-out LP collection into my iPod, I'd have no issues with plugging a
> cheap preamp into my soundcard and using Adacity. If I were a professional
> trying to preserve an archive of unique sound recordings, I'd scale up
> substantially and do the job right so I didn't have to do it again and I
> could leave a legacy of high-quality digital audio for the future.
> I'm sure any of the pro-grade programs are in the same league as Soundforge,
> btw, I just recommend because I use it and know it. I will say that I find
> Protools overly complex for almost anything I'd do, and a
> tracking/sequencing program like Acid or Cubase may or may not be
> appropriate for this sort of work. Its seems like DC6/DC7 would work just as
> well, too, and perhaps at a lower cost.
> Finally, let me emphasize that software choice should be secondary to as
> good an analog playback as possible and as good a digital interface as
> possible. The hardware quality or lack thereof can ruin a transfer effort
> right from the start. With any competent software used by a person who knows
> how it works, you have to work at it some to ruin the audio quality. It's
> more a matter of how you want to spend your time -- fussing with primative
> and klunky interfaces or getting the job done efficiently.
> -- Tom Fine
> ----- Original Message ----- From: "Sam Brylawski" <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Thursday, August 14, 2008 10:19 PM
> Subject: Re: [ARSCLIST] Seeking recommendations for oral history
> digitization equipment (fwd)
> You've made an unequivocal non-recommendation of Audacity. For someone
> on a budget who is only re-formatting/migrating, presumably w/o
> "processing," exactly why shouldn't they use Soundforge? Deatils would
> be very useful.
> On Thu, Aug 14, 2008 at 9:37 PM, Tom Fine <[log in to unmask]>
>> The most cost-effective way to do a large-ish transfer project is do it
>> and do it right. So, you don't want to cut too many corners. Joel has laid
>> out a bunch of good points before. Even for cassette-quality oral history
>> type stuff, it's best to do a good -resolution transfer once on a
>> well-maintained deck into a decent digital signal chain. Bottom-barrell
>> stuff will give bottom-barrell results but there is a whole range of
>> reasonably-priced and good quality gear out there.
>> As for software, I hate Audacity. I think it's worth exactly what it
>> It's klunky garbage, at least the last version I used, which was 2 years
>> ago. Much better is Sony Soundforge which can be had very reasonably if
>> is an education-related institution (see Educator Superstore website for
>> instance). The newest version of Soundforge comes with restoration/cleanup
>> tools that, when used conservatively and tastefully, can be very helpful
>> with this kind of audio.
>> Your best practice is to plan on at least three collections of files: 1)
>> transfer PCM audio, hopefully higher resolution (I like 88.2/24 or 96/24
>> Richard Hess makes a convincing argument that 44.1/24 is perfectly OK for
>> spoken-word material). 2) processed PCM, this would be NR'd, normalized,
>> EQ'd etc and perhaps saved at CD resolution with a "safety" version burned
>> to archival CD media. 3) online/small-format version, MP3 or whatever
>> crunched format you used, saved from the CD-quality processed PCM version.
>> This would be for online/streaming or podcast use. These can be batch-made
>> by Soundforge out of the processed PCM files. My caveat would be, beware
>> crunching too lossy. Spoken word starts sounding really crappy when it's
>> surrounded by digi-swishies and other artifacts. I never go lower than
>> 96kbps for MP3, which some might consider overkill but I sure don't. I
>> actually prefer 128kbps whenever possible because it preserves the upper
>> mouth/throat air resonances of most voices and also a decent MP3 cruncher
>> won't make swishies out of even heavy tape hiss.
>> Don't neglect the cassette end of this. Having a Tascam 122MkII is one
>> but, how old is it and how well-maintained is it? I highly recommend
>> it to NJ Factory Service for a refurb and have him make sure the head is
>> too. As of a few months ago, Tascam still had heads, belts and other parts
>> for these machines.
>> Now that you see that even this kind of "pedestrian" audio is no simple
>> to transfer and preserve correctly, have you considered out-sourcing the
>> transfer work? You could then concentrate your expertise on archiving,
>> editing, and making available your assets. There are grants out there to
>> for outsourcing audio work to audio professionals.
>> -- Tom Fine
>> ----- Original Message ----- From: "JA Eaton" <[log in to unmask]>
>> To: <[log in to unmask]>
>> Sent: Thursday, August 14, 2008 12:11 PM
>> Subject: Re: [ARSCLIST] Seeking recommendations for oral history
>> digitization equipment (fwd)
>> Hi John,
>> Some thoughts on your situation...
>> 1) As you are digitising for the first time, it's worth thinking about
>> preserving your files for future use, especially if you expand this
>> resource in the future. Therefore it's probably worth digitising your
>> material in the highest quality available in a format best suitable for
>> sustainability in the best possible way.
>> Which leads to...
>> 2) Record your audio in the best available quality and convert down to
>> CD/web quality afterwards. i.e. 88.2Khz/24bit. You can back up the
>> files onto a hard drive or DVD's/CD's. Even though you may not need
>> these high quality files now, it'd save any future re-digitising issues
>> provide you with a backup if anything goes wrong. You may wish to consider
>> using open source file formats for future proofing such as Vorbis ogg or
>> FLAC, whereas WAV or AIFF are fine for CD distribution.
>> which leads to...
>>> One of the USB interfaces I was looking
>>> at (the Tascam US-144) comes with a free version of Cubase, but I don't
>>> know that it would be any better for our purposes than Audacity
>> A dedicated A>D converter over USB 2.0 or Firewire is going to give you
>> much better conversion quality than plugging into your internal soundcard.
>> These vary in quality and price and will generally be defined by your
>> budget. Quality of the pre-amps you use is also worth noting.
>> 4. Pro Tools is a highly advanced multitracking and editing suite which
>> although capable is probably far too advanced for your needs. If you were
>> planning on digitising multiple media at once (multitracking) then
>> something of this ilk may be worth thinking about. (note that Pro Tools is
>> only compatible with it's own hardware except for Pro Tools M-Powered, a
>> lighter version). Audacity on the other is a freeware simple interface for
>> recording, simple editing, basic processing techniques and file
>> One consideration is that the simpler (and cheaper) the program then the
>> less 'restoration' processing features it is likely to offer (should you
>> need them), such as de-noisers etc.
>>> Anyway, your thoughts on a good USB interface, a good and inexpensive
>>> pair of monitors, headphones
>> a good pair of monitors may be worthless if your listening environment is
>> not designed for audio analysis(e.g. your desk is in a big open plan
>> with lots of background noise), however there are some decent reference
>> monitors on the market under the £400 mark. (KRK, Genelec being at the
>> end of the scale).
>> Again look for professional headphones with a 'flat' response (i.e. not
>> marketed to D.J's, live broadcast etc) but this can often be tricky
>> between brands. Try AKG, Sennheiser.
>> USB interfaces. at the top end of the price scale (for your project
>> anyhow)MOTU make decent converters with quality preamps, alternatively
>> M-Audio, Edirol and Mackie do cheaper products. Look for ones which offer
>> higher sample rates/larger bit depth for any future needs.
>> hope this is of some help!
>> Joel Eaton
>> Joel Eaton TSO - Sound Resources
>> TASI - A JISC Advisory Service
>> [log in to unmask]
>> Free Helpdesk for UK Further and Higher Education: <[log in to unmask]>
>> Online advice documents: <http://www.tasi.ac.uk/>
>> Hands-on training: <http://www.tasi.ac.uk/training/>