Skip to main content

Hey guys!

If you're not already familiar with this site, it's an awesome one to keep bookmarked. It has more vacancies posted than International Musician!

BTW, there was a great (depending on view point) article posted in their forums that I just felt the need to reply to. Feel free to check it out.

http://myauditions.com/eve/ubb.x?a=tpc&s=217607196&f=2256096313&m=621108321&r=632107621#632107621

J.

Comments

DavidSpearritt Fri, 09/23/2005 - 14:38

Here's what I think will work.

1. Maintain a registry of MA certified engineers, ie ones who pass certain basic prerequisites, ie can read a score, are human beings, can communicate etc, get this available to artists, on the web, whatever. In the registration, allow the engineers to specify a non-exclusive interest, ie orchestral, chamber, opera if they want. Display this "interest" to engineer hunters when they search.

2. Maintin a feedback system ala eBay, for artists and MA admin to submit to after the recording. Allow this feedback to be shown to future artists when checking available engineers.

Forget exclusivity for region, this will never work and is probably against the law, forget any technical specs, leave that to the engineers, forget hall definition, any hall agreed by the engineer will do. If she is good, she will know which hall and mic technique to use given the ensemble details. One size fits all mic and hall choice regardless of ensemble makeup will never produce optimum results.

Available feedback rating is the key.

anonymous Fri, 09/23/2005 - 17:02

I wasn't really making a joke, but I'm not offended if folks think the idea is amusing, nor even if they think it's stupid. It's an idea for discussion, that's all.

It wouldn't take God's Own Encryption to make this "magic box" protocol a lot less susceptable to cheating than anything discussed so far. There are readily licensable algorithms for DES and public key encryption. It's only necessary to pick something that could be done in real time for small packets. The code need not be very robust if it is changed periodically, and if MyAuditions retains custody of all the hardware.

As for the 44.1 vs. 96 kHz problem, you can do 44.1 if you can do inaudible watermarking at a sufficient bit rate to detect edits. I don't accept that watermarking would yield unacceptable audio quality FOR THIS APPLICATION. While I personally believe I have detected watermarking in certain situations, I wouldn't want to bet very much on passing a double-blind test of my ability. Orchestra audition committees used to put up with really terrible cassette tapes. They are not going to care about watermarking.

I didn't propose sub-audible watermarking, simply because I wasn't sure it could be done in real time on a single DSP. I was trying to propose something that I knew for certain I could build. The supersonic "audit signal" I suggested was chosen for ease of implementation. To make it withstand the "spectral splitting" attack I described earlier, this audit signal must include some low-bit-rate coded audio. No problem: such encoding algorithms are readily licensable and can run in real time.

I think you guys are not listening carefully to the client. What he WANTS is good quality audio. What he NEEDS is a guarantee of no edits. I've proposed a way of providing the later without precluding the former.

Such a "magic box" is not just blue sky. I could certainly build one in a year, given proper funding. Of course I already have a day job that I like perfectly well, and it would take some serious incentives to make me leave it. But I could absolutely build this, and what I can do, someone else can do too.

David L. Rick

anonymous Fri, 09/23/2005 - 18:31

OneMegahertz wrote: I think you guys are not listening carefully to the client. What he WANTS is good quality audio. What he NEEDS is a guarantee of no edits. I've proposed a way of providing the later without precluding the former.

And an elegant solution it is! But how much easier can you get than handing the client's representative CD-R of what was just recorded?

Rich

DavidSpearritt Sat, 09/24/2005 - 17:49

And an elegant solution it is!

Not! This sort of complexity will never work, the KISS principal needs to be applied.

The CDR on the day is a great solution, but I will want to have applied small amounts of compression, EQ and possibly conv reverb, to the recorded stream going to CDR. This is required to get the best result in all but the most superb acoustic with qualiity performers. Most audition conditions are far from this on both counts.

I think MA could really contribute in creating a great directory of personnel and contractual conditions (no edits) rather than getting involved in techo issues.

anonymous Sat, 09/24/2005 - 19:11

DavidSpearritt wrote:

And an elegant solution it is!

Not! This sort of complexity will never work, the KISS principal needs to be applied.

The CDR on the day is a great solution, but I will want to have applied small amounts of compression, EQ and possibly conv reverb, to the recorded stream going to CDR. This is required to get the best result in all but the most superb acoustic with qualiity performers. Most audition conditions are far from this on both counts.

Perhaps in Australia "elegant" and "simple' are synonyms but not here. It seems strange you are pleading for simplicity and then announcing plans for compression, EQ, and reverb!

Rich

DavidSpearritt Sun, 09/25/2005 - 05:02

Yes, I suppose you are right, I do believe elegant and simple are synonyms.

But I think a little EQ, reverb and compression is not that hard, I use a Gold Channel to do this when recording the "CDR on the night" for the visiting artists. Most often our naked 24 bit file does not sound that great going to CD direct for the following reasons:

1. The level is too low, as we record cool at 24 bits to get headroom for the unexpected.

2. In normal to bad acoustics we err on the side of being slightly too close, so reverb can repair it, too far away and you are stuffed.

3. A tiny amount of compression (not always) helps lift some low output performances, which in auditions is quite common.

Going digital through one box for these is quite simple really. Mucking around with supersonic watermarking signals and preventing media transfer with CD audio is not. :)

alexaudio Sun, 09/25/2005 - 23:07

Hello everyone,

I have been perusing this topic for a while. The mission of My Auditions is an honorable one and in my minds eye, worth pursuit. I have also taken a brief look at their website, which provides lots of useful information for the classical music scene, educational institutions, etc. – a great resource.

The topic at hand has brought up many concerns regarding guidelines and procedures on the best manner to capture applicants for music schools, orchestras and the like. The mission of My Auditions, as it seems, is be able to provide orchestras, music educational institutions and others with a fair and impartial recordings so that those institutions can then make choices on whom to further pursue for their selection process.

To setup guidelines and procedures for this endeavor is not all that dissimilar than setting up test methods for comparative evaluations of equipment, such as Reference AB tests, MUSHRA testing and other tests that allow subjective judgments to be made without imposed bias. Of course, this topic has many more variables than a controlled test. However, the same methodologies that are utilized to set the standards and guidelines for the aforementioned tests can be applied here (including technical and psychological).

Having recently completed the construction and administering of an ITU-R Reference-AB surround sound test a few weeks ago – I see a lot of parallels in appropriately setting up the guidelines and procedures for My Auditions endeavors as well. It took us (AES, NPR along with educational institutions and testing specialists that are knowledgeable in ITU-R guidelines) weeks to construct the appropriate testing guidelines and procedures. I see that this will be the case for My Auditions as well. Jack Reynolds of My Auditions comments on forming a “technical committee.” I believe this is absolutely appropriate. Though committee business sometimes elongates the decision process; it is appropriate in this situation so that the appropriate dialog exists.

To have merit and credibility, I hope that the committee utilizes similar standards to that of the National Radio Standards Committee (which I am a member). Using the NRSC as an example – I would hope that the committee’s formed utilize appropriate meeting guidelines (such as Rodgers Rules) and has two committee chairs that are aware of and have a history of conducting functional committee meetings. Of course an appropriate secretary would be chosen to document what transpired in the meetings. It should be known that many of these type of committee’s are NOT formed by an electoral process, but are appointed. (However, these committee’s do vote upon the items discussed.) Just saying this in case if any of us are requested, or not, to participate on the committee, that we should NOT take it personally.

My personal recommendation (keep in mind these are only suggestions), which come from someone who has administered and is currently participating in effective standard setting committees, is that the committee members for the audio engineering technical committee would compromise of audio recording engineers who are:
a) Knowledgeable and have certified education (or equivalent documented experience) in recording classical music,
b) Have a recent work history with professional musicians applying for auditions similar to those that My Auditions represents,
c) Have a current working relationship that is easily documented and/or with referrals for orchestra’s and/or educational institutions that My Auditions represents.
d) Understand many of the technical and scientific background of audio recording ‘engineering,’ as well as the artistic merit of the audio profession.
Of course, since My Auditions seems to be a private institution, none of the recommendations I and/or anyone else may be taken. That has to be understood so we don’t have a ‘heated’ conversation where emotions flare (which I have seen a little of on here). I would imagine that it is a possibility that committee members may also have to sign a NDA (non-disclosure agreement).

Being an engineer that works with one of the most recorded orchestra’s in the America’s on a regular basis, as well as with Musicians Local No. 1 on a regular basis and like others on this forum, with a high degree of technical and artistic expertise in recording classical and acoustic music - I also see the great importance of creating guidelines and procedures that can be applied that:
a) Provide a benefit to the musician, knowing that their recording allows them to be captured in the best manner possible,
b) Provides the client (My Auditions) with material that has minimal bias applied (i.e. edits),
c) Allows the audio recording engineers to perform their task at hand to provide quality results,
d) Allows all parties a fair and equitable means to make a living (financially speaking).

That all said, it has to be understood that, among the recording professionals in this discussion, that audio recording of music is an art form and a science. To have the recording process to be as impartial as possible, we need to take out the art form and our personal bias out of the conversation momentarily. Then we can look at the variables from an engineering/technical/science point of view that can be controlled and put into a contractual agreement that My Auditions’ engineers would have to sign. Many of these variables really won’t be that hard to identify Things such as standardizing guidelines/procedures such as:
-average RMS level of all submissions
-type of coder and bit rate for submittal
-a choice of recording techniques that would allow the majority of engineers to perform an effective recording.
-no editing
-no limiting/clipping for most submissions
-extent of compression (like any one of us would use 6:1 compression anyways)
and the like.

I have scene many recommendations on here that are quite well founded. The last, which is around watermarking, is an excellent idea that could actually work. One of the surround sound composite encoder systems I am working with utilizes a watermarking technology. It is incredibly robust. To utilize a watermark in this scenario - it would have to be a watermark that utilizes a COFDM structure to remain robust and be able to pass through manipulation and codecs. However, I am not sure it would withstand the additions of reverb…though it is easy for me to find that out. I know that it can withstand EQ and heavy compression quite easily and even multiple cassette duplications (because I have tested it). It would not need an encryption code. All it would have to be is a time code data. If the time code data shifts in the middle of the performance, bingo, you have evidence of an edit. All other methodologies discussed to detect edits can be proven unreliable. The only problem with the aforementioned watermark is that this watermark can only be applied digitally, after ADC. This would be difficult to manage, since the watermark could be applied anywhere in the digital chain, even after edits. However, the watermark could be utilized to certify the authenticity of the recording (when, where and who it was). This, I think would be more valuable, as then the recording can be given to My Auditions and the to musician(s). If copies are made and get into the wrong hands, it would be easy to detect who the artist was, where they performed, etc. The decoder would be simple and easy for anyone with a computer with a CD-ROM to utilize.

Cucco Mon, 09/26/2005 - 05:21

Okay, so above are some great suggestions. Here are a few comments.

David/Rich -

I agree with David here. Simple and Elegant are (or at least should be) one in the same. A simple signal chain allowing the effects to make the sound as good as possible without performing physical (or logical) edits is a great view of simplicity and elegance.

(BTW, why the slam on the land down under??)

Perhaps in Australia "elegant" and "simple' are synonyms but not here.

Alex -

Standardization is also an area which (unfortunately) I'm infinitely familiar. My "day job" with the government involves me sitting on many standards panels. The one thing that you really hit on here which is true is that the art must be taken out of the process if there is truly to be some kind of standard for the sound. This is an unfortunate truth. An even more unfortunate truth is that, this will automatically give others a "leg up" on the audition rounds because they will be able to get an engineer who is not bound by these standards.

So, to me the only possible solution that remains is to not standardize the recording practice at all, rather set forth some rules which MUST be abided by for an engineer to remain in good status with MA. Of course, the "no edit" rule would be high on the list. Perhaps things such as a required MPEG codec (supplied by MA), sample rate/bit rate and so on are still well within the science side of recording that these things could be specified without any weakening of the integrity of the recording.

Personally, I think the watermark idea, while creative and thought-provoking, would be a huge head-ache and would not likely yield the return on investment. In the military, when we create a solution to a problem, the problem must have some merit or real threat before anyone will throw any money at. In other words, if edits are an issue which can be subverted in a less-expensive means or an issue that, with contractual policy, can be all but avoided completely, a costly encryption.

If we're looking for a non-repudiation (eg. "I'm the recording engineer who recorded it," and "I'm the auditionee" and "I'm the proctor") a simple biometrics-based system would do the job quite well. (Such as DigitalPersona's U-Are-U Pro system - at roughly $150 per desktop. In other words, each engineer has a reader which the engineer, performer and proctor can all register and submit a biometric record along with the recording.)

More $.02

J.
[/img]

alexaudio Mon, 09/26/2005 - 08:13

DavidSpearritt wrote: This is a very pertinent summary of the issues facing the proposed scheme. Playing the devil's advocate again, I wonder how much time MA have got. Alex, how long do you think all the issues would take to finalise? Also, the problem of watermark timecode reprinting during mastering is something of an unknown.

If MA (I am going to use that abbreviation now for My Auditions) formed the tech panal and it is not to large or a group...I would recommend no more than 9 (5 or 7 preferred), I think the entire proposed scheme could be done in about 2-3 months tops. If watermarking gets into play, that is about the only thing I can see that could really elongate the process. The way that the committee meetings I serve on (with the NRSC and others), meet via teleconference. I would do the same for this one.

alexaudio Mon, 09/26/2005 - 08:24

Cucco wrote:

Alex -

Standardization is also an area which (unfortunately) I'm infinitely familiar. My "day job" with the government involves me sitting on many standards panels. The one thing that you really hit on here which is true is that the art must be taken out of the process if there is truly to be some kind of standard for the sound. This is an unfortunate truth. An even more unfortunate truth is that, this will automatically give others a "leg up" on the audition rounds because they will be able to get an engineer who is not bound by these standards.

Agreed, however, I believe that if the standards committee were to go farther than MA and put this in place for the union, symphony orchestra league, etc....that could really reduce the likelyhood of other engineers having a "leg up." Just matters on the overall plan of attack.

Cucco wrote:
So, to me the only possible solution that remains is to not standardize the recording practice at all, rather set forth some rules which MUST be abided by for an engineer to remain in good status with MA. Of course, the "no edit" rule would be high on the list. Perhaps things such as a required MPEG codec (supplied by MA), sample rate/bit rate and so on are still well within the science side of recording that these things could be specified without any weakening of the integrity of the recording.

Personally, I think the watermark idea, while creative and thought-provoking, would be a huge head-ache and would not likely yield the return on investment. In the military, when we create a solution to a problem, the problem must have some merit or real threat before anyone will throw any money at. In other words, if edits are an issue which can be subverted in a less-expensive means or an issue that, with contractual policy, can be all but avoided completely, a costly encryption.

If we're looking for a non-repudiation (eg. "I'm the recording engineer who recorded it," and "I'm the auditionee" and "I'm the proctor") a simple biometrics-based system would do the job quite well. (Such as DigitalPersona's U-Are-U Pro system - at roughly $150 per desktop. In other words, each engineer has a reader which the engineer, performer and proctor can all register and submit a biometric record along with the recording.)
[/img]

I don't believe that this would be a costly encryption. The tech. is already available and would just have to be licensed. I believe it would be a head-ache to prove that there were no edits, but very appropriate to identify who's recording it is. I wouldn't use "DigitalPersona's U-Are-U Pro system" as it is much more complex to manage and would likely cost more than a simple watermark software plug.

Cucco Mon, 09/26/2005 - 10:16

alexaudio wrote: [quote=Cucco]

Alex -

Standardization is also an area which (unfortunately) I'm infinitely familiar. My "day job" with the government involves me sitting on many standards panels. The one thing that you really hit on here which is true is that the art must be taken out of the process if there is truly to be some kind of standard for the sound. This is an unfortunate truth. An even more unfortunate truth is that, this will automatically give others a "leg up" on the audition rounds because they will be able to get an engineer who is not bound by these standards.

Agreed, however, I believe that if the standards committee were to go farther than MA and put this in place for the union, symphony orchestra league, etc....that could really reduce the likelyhood of other engineers having a "leg up." Just matters on the overall plan of attack.

Now your hitting on something! If the whole process were standardized, then it would be a win-win for everyone involved. That would truly be a revolutionary movement and would take great effort to get through, but in my humble opinion, quite worth it!

Cucco wrote:
So, to me the only possible solution that remains is to not standardize the recording practice at all, rather set forth some rules which MUST be abided by for an engineer to remain in good status with MA. Of course, the "no edit" rule would be high on the list. Perhaps things such as a required MPEG codec (supplied by MA), sample rate/bit rate and so on are still well within the science side of recording that these things could be specified without any weakening of the integrity of the recording.

Personally, I think the watermark idea, while creative and thought-provoking, would be a huge head-ache and would not likely yield the return on investment. In the military, when we create a solution to a problem, the problem must have some merit or real threat before anyone will throw any money at. In other words, if edits are an issue which can be subverted in a less-expensive means or an issue that, with contractual policy, can be all but avoided completely, a costly encryption.

If we're looking for a non-repudiation (eg. "I'm the recording engineer who recorded it," and "I'm the auditionee" and "I'm the proctor") a simple biometrics-based system would do the job quite well. (Such as DigitalPersona's U-Are-U Pro system - at roughly $150 per desktop. In other words, each engineer has a reader which the engineer, performer and proctor can all register and submit a biometric record along with the recording.)

I don't believe that this would be a costly encryption. The tech. is already available and would just have to be licensed. I believe it would be a head-ache to prove that there were no edits, but very appropriate to identify who's recording it is. I wouldn't use "DigitalPersona's U-Are-U Pro system" as it is much more complex to manage and would likely cost more than a simple watermark software plug.

Well, as for the encryption - if you're using a standard DES or a 3-DES, what would be the source of the keys? True, it wouldn't be expensive, but there are many "questionable" sources of encryption schemas. The most robust and trustworthy are those that have been evaluated and certified by the government(s) (Such as FIPS 140-2 Certification which is the NIST standard which certifies all crypto modules for use on US Government networks).

As for the digital persona option, this isn't difficult to set up in the slightest... It's real simple:
1. the Main software package is loaded somewhere at MA headquarters
2. Each field engineer is issued a fingerprint reader and the DP software (which requires only minimal software and is advantageous on other levels - system security and password management).
3. For each recording, the necessary parties place their index finger on the platen (3 times for accuracy - all told about 12 to 15 seconds for each person) and then you store the biometric template and associate it with the individual file. A process which is all but fully automated.
4. When you burn the disc image, the templates are then affixed to the disc and sent to MA.

Upon receipt of the biometric template, MA could
1. verify that it was in fact one of their approved engineers submitting the recording.
2. Store the auditionee's template so that if an orchestra invites one of these players to an audition, they can in turn whip out their DP scanner and verify that the person who performed on the CD is in fact the same person standing before them.

And, since the DP software (as with most other biometrics software - but not all, so beware...) stores the files as templates only (not images), privacy rights are left in tact. As well, they are stored as encrypted hashes and the transmission is encrypted from point to point (with a FIPS 140-2 accredited crypto-mod - your choice - AES/3-DES/Blowfish), you don't have to worry at all about "fingerprint theft." (Which is impossible anyway since an image cannot be reconstructed from a template).

To me, this appears rather simple - far more than devising an encryption scheme to watermark discs (though they do serve two slightly different purposes.)

J.

Zilla Mon, 09/26/2005 - 11:44

Cucco wrote: ...the necessary parties place their index finger on the platen (3 times for accuracy - all told about 12 to 15 seconds for each person) and then you store the biometric template...

Maybe you are just offering this as fuel for thought or maybe I misunderstand, but....

There is NO WAY I am offering up my fingerprint to an on-line database. This would expose people to indentity theft and raises serious privacy concerns. I certainly would decline to participate in any such program.

Cucco Mon, 09/26/2005 - 16:31

Zilla wrote: [quote=Cucco]...the necessary parties place their index finger on the platen (3 times for accuracy - all told about 12 to 15 seconds for each person) and then you store the biometric template...

Maybe you are just offering this as fuel for thought or maybe I misunderstand, but....

There is NO WAY I am offering up my fingerprint to an on-line database. This would expose people to indentity theft and raises serious privacy concerns. I certainly would decline to participate in any such program.

Hey Zilla -

This is simply not true and not possible. It's a myth created by some political lobbyists and perpetuated by hollywood.

Here's how biometric technologies work (virtually ALL, with the exception of the FBI's IAFIS system and a few DOD systems used for terrorist tracking) -

An image is scanned, however, it is immediately turned into a binary string known as a template. No image is even so much as cached on the machine. The template is made by examining minutae points on the fingerprint itself and using that, based on proprietary and protected algorithms, turns those minutae points into a data string. Since the scan can be of different part of the finger and the quantity of minutae points (as well as their locations) can vary quite a bit, even if someone were able to "reverse engineer" the binary string into an image, it would not be an image of a finger print. It would simply be the appearance of a constellation and nothing more. And, since the minutae points are not identified within the code as to their type, it would be virtually impossible to determine what type of minutae point it is.

There is no "pie in the sky" database of all of our fingerprints which someone could hack and use to steal our identities. The only thing that comes close would be the FBI's IAFIS system and since I have visited this system numerous times, I can assure you that it is QUITE secure. I can also assure you that you aren't in it (unless you have either obtained a security clearance or have commited a federal crime - or if you're on a terrorist watch list)

The data which would be stored (in an "encryption at rest" schema) would simply be a string of 1's and 0's. Furthermore, since the systems have "failsafes" in them, if someone were clever enough to decode the 128 bit hash of the data and tried a "replay" attack, the system would recognize the 100% match and reject it immediately b/c it is aware that this simply is not possible.

And, just in case you doubt my credibility on this one, here's the signature block from my work e-mail address:

Jeremy Cucco
US Army Biometrics Liaison
XXXX Crystal Drive
Suite 7136
Arlington, VA 22202
703-602-XXXX
jeremy.cucco@us.army.mil

(I have XXX'ed out some of my address info and telephone info as this is not an appropriate forum for me to disclose this information.)

BTW - for a further explanation of my position - I am the Army's sole liaison to the DoD Biometrics Management Office. I work out of the Pentagon and am the Army's subject matter expert on policy, science and implementation of biometrics technologies. Of course, fingerprints are one of the more common biometrics identifiers, but I'm sure you are well aware that there are many others.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~``

BTW - If I wanted to steal yours or anyone else's finger prints, all I'd have to do is go behind you at a restaurant with a Boy Scouts grade finger print kit - lift the print and have a rubber stamp made of the image. Ironically, this will fool many of the lesser sensors on the market, however, there are many sensors which include some type of "liveness detection." Much of this information is classified, so I obviously will not go in to it here. However, imagine this:

If I don't like someone, I can easily do the above and "make your fingerprint." I can then take the rubber stamp and place it on my forehead (to get some "human grease" on it) then place the latent print all over a crime scene. You are now suspect # 1 if you happen to have any connection to the victim. (Cuz believe me, if you know the "victim" you're gonna get printed.)

Needless to say though, it's far easier for me to steal your identity this way than to hack a 128 bit database of binary code which would need to be reverse engineered.

So, I hope this helps to clarify some of the misconception.

Jeremy :D

Zilla Mon, 09/26/2005 - 17:30

Cucco wrote:
There is no "pie in the sky" database of all of our fingerprints which someone could hack and use to steal our identities. The only thing that comes close would be the FBI's IAFIS system and since I have visited this system numerous times, I can assure you that it is QUITE secure. I can also assure you that you aren't in it (unless you have either obtained a security clearance or have committed a federal crime - or if you're on a terrorist watch list)

Actually, I did obtain a classified security clearance back in my Naval Nuclear Power days. It is understandable to give up your fingerprint knowing that national security is in the balance, and that a very secure and regulated government entity (FBI) is in custody of my personal data. While no database has perfectly impenetrable security, it is reasonable to believe that the FBI is very unlikely to be compromised.

MA is not a regulated government entity. It is a private company based publicly on the internet. With due respect to your expertise on the matter, you will not be able to convince me that this course of biometric certification is advisable. Maybe I will be perceived as an alarmist. Your clear explanation does make the risk seem small. None the less, I am simply not willing to expose myself to that kind of risk for a few grand of extra income per year.

Cucco Mon, 09/26/2005 - 18:55

I understand. Unfortunately, the media portrays the fingerprint issue as something easily subverted by the removal of ones finger and a database that is infinitely hackable.

The great news about a system like I purposed is that it is certified for government use, yet affordable and within reach of the average consumer. While I think your hesitation, personally, is unfounded - you are certainly entitled to it.

Of course, be aware that, in the very near future, you will be supplying a biometric of some sort to a computer near you. (passports, job applications, building access, airport access, etc.)

(Actually, since you're in CA, I can assure you that, whether you're aware of it or not, your biometric information has already been collected and is being used for authentication purposes -- think DMV...)

J. 8)

recordista Mon, 09/26/2005 - 19:34

Wow--been away for awhile and just found this incredibly juicy thread.

As someone who has been recording audition tapes (on and off) since the early '80s, I applaud the effort to make unedited audition tapes available to potential employers or grantors. As a technologist with two plus decades of experience in television, telephony, and Internet operations I understand that this is impossible to truly guarantee. The watermark idea is interesting, and could prevent editing of any kind -- but would require high samplerate archives of every recording which were used to generate the online files. I think it's overambitious right now, but might be an interesting future improvement -- particularly if evidence of edited recordings surfaces after the first year or so.

Specifying particular mics and their placement seems to me a spectacularly poor idea. Most of us have chosen our tools based on how they work together and have become familiar with the performance of the ensemble as it were -- combine this with the already discussed differences in venues and you are headed for a real train wreck.

I think the technical committee is a good idea--but I share the concerns of others that this could degenerate into a process-oriented nonproductive exercise.

Please do keep us advised of how this shakes out.

alexaudio Mon, 09/26/2005 - 19:37

Thank you Jeremy for the detailed explanation of the Biometric system. I can back Jeremy up on this one that it is a very safe and secure system. It could be used by MA or any other business or institution. A similar system is being test marketed in the western US for buying groceries of all things. At this point, there has not been any loss of financial and/or personal information on the systems being test marketed. They are very robust and extremely secure.

However, for the application at hand, they simply are not appropriate. It is out of the scope of discussion and would not be needed for MA application. The amount of added tech, software, hardware and cost would be an extra burden to an administrator that simply isn't necessary. A watermarking scheme that could be provided for a small licensing fee and as a software plug via download (say for $35) would cost far less and be easier to manage. There are more major concerns about the debate at hand than the security of the material. Audition tapes are not a matter of high security....just authenticity, which could be handled in a much more simplistic manner.

ghellquist Mon, 09/26/2005 - 22:42

I have been reading this thread partly with interest and partly a bit taken off. Beeing only an amateur recorder, maybe I should keep my mouth shut, but this is getting a bit out hand, isn´t it?

I mean the whole purpose is to make a decent recording of a person, allowing a jury to decide which persons to actually call to the interview. We all know that this is most often only the very first step in a long process. No orchestra I could think of would hire on this material only, they would want to meet the person and interact with the personality. And I guess the first thing the jury would ask the player to do, would be to play the same pieces as on the recording. In real life, no edits.

So, I would actually keep it rather simple. If a simple scheme doesn´t work, then you can later make it complicated.

- the recording engineer fills in a form, stating how the recording was done. It might have tick marks for how long time it took to do it, a tick mark saying that no edits was done, what equipment used. Also a note on what identity card (or whatever) the talent did show. And a signature in writing from both the engineer and the talent. Add to that a digital photo of both the talent and the room.
- now the recording engineer could upload the file to MyAudition, and send the filled form as well.

This would rely on the honesty of the recording engineer as well as on the honesty of the player. But then again, any scheme we make up can be cheated if you really want to.

Gunnar

DavidSpearritt Tue, 09/27/2005 - 02:07

Yes Gunnar, it has got way out of hand and over the top, with all this stuff about biometrics. The cost, time involved to set up and the unecessary complexity with such a scheme would ensure its failure. Modern life is getting expensive and messy. It could even turn into something like this ...
http://news.scotsman.com/index.cfm?id=1965122005
So much for the hi-tech solution.

Re-reading my posts in this "juicy" thread, I humbly apologise for being blunt and cynical.

Cucco Tue, 09/27/2005 - 04:39

Hey guys!

I know, the Biometrics idea is a bit "over the top." I only bring it up cuz I know the systems inside and out - what they are capable of and how easy they are to use and implement.

Many people have now hinted at the complexity of this system and how it would be ineffective. And it's not my point to belabor this issue - only I want to add that the complexity is actually non-existent. These systems are so easy to install, even my own mother has done one. (And that's saying something... :-? )

As well, many of these systems have encryption tools and watermarking templates built in already. It's merely a suggestion (with a motive -- yes, I do have an agenda here - I'm doing my damndest to get the public aware of biometric systems, their positives and exposing the myths. So much so that I was actually scheduled to meet with Chris Alexander - a Fox TV Studios executive and possibly even Kiefer (sp) Sutherland to portray the DoD's unclassified uses of biometrics on the television show 24. It was only after one of the folks in my reporting chain had a problem with one of the items did the whole thing get called off.)

But, then again, I was also thinking on a slightly more grand scale than just the audition/recording process too. We do want to make sure that only certain people can access these recordings on the back end.

IOW -

Orchestras (or more specifically, managers within the orchestras) can only listen to recordings

Users can only listen to their own recordings and remove existing recordings

Engineers may be able to upload recordings

And so on. So, ultimately we are dealing with a very large, enterprise type system with role-based and rule-based access.

I know, I know I'm being a DORK. That's just me.

I think the reason this topic has gotten so "wild" or diverse is because everyone here is actually quite interested (either positively or negatively) in this system. This isn't a bad thing - just interesting.

Dave - Don't beat yourself up too much over being blunt and cynical (though, you're probably not... :P ), often it's the blunt and cynical who make us open our eyes. Of course, being a blunt and cynical person, you're also aware that when one's eyes are opened, they usually don't like it. (Just try opening my wife's eyes at 5 o'clock in the morning to feed the baby!!!)

J.

Thomas W. Bethel Tue, 09/27/2005 - 04:55

Yes there is a problem with people wanting to appear to be what they are not and yes they want to present themselves in the best possible light for grad schools or orchestral auditions.

The problem with what is being discussed here is that it sounds more like preventing terrorism than the standardizing of recordings of audition tapes.

If you want to cheat there are was to do it and there is someone can be paid enough to help the person cheat.

Lets face it the competition for jobs and grad schools is intense and there are lots of people who would do almost anything to get a good life long job with a major orchestra or get into one of the top grad schools. This would include editing together multiple takes so that they sound like the musician they want to be rather than the musician they are.

No matter how many rules or how much hi tech mombo jumbo you have someone will figure out a way to get around it.

I applaud Audition.Com for what they are trying to do but I think it is virtually impossible to do what they are trying to do because you have the "human" factor to consider.

Most classical recordings that are released today contain hundreds if not thousands of edits so what the buying public is hearing is a patched together work that did not exist in one piece. Is it any wonder that young musicians seeking a job or space in a grad school would want to do the same thing when they are making their audition tapes? They read the same stories as the rest of us and find that in the age of "perfection" and the advances in digital editing make it possible that every single note of a piece could be edited to make it perfect.

The only real way to do all of this is either an "in person" audition or to video tape the audition with a setup camera in one spot and time code running on the screen so that there can be NO editing of the audition since any edit would have to be concerned with both the audio and visual and if you had time code burned onto the screen any attempt to edit the music would upset the time code displayed on the screen. This would also provide the grad school or orchestra the ability to see where and when the audition was taped and how it was miked and how the person played. If they don't want to consider the person based on any physical characteristics they could turn off the video and just have the jurors listen to the audio thereby protecting the identity of the player and having the jurist concentrate on the aural activity.

I am sure there would be ways to cheat this way as well but the chances are much greater that the cost and the technical problems would prevent anyone from easily doing this. With most DV cameras there is digital stereo audio and would be almost as good as providing a CD of the audition.

This has been a very enlightening discussion and has touched on matters I never really considered in doing audition tapes for people seeking employment or entrance into grad school. This is a GREAT forum.

DavidSpearritt Tue, 09/27/2005 - 05:30

Tom, I think you have nailed it.

I don't know why I didn't mention it before, but most of our audition recordings are in fact dual system video and audio because recent orchestral position and opera school entries are requesting it, for obvious reasons.

Because dual system video and audio is much more difficult to edit, (and a real pain with most currently available software) you have a much better chance of maintaining data integrity.

JoeH Tue, 09/27/2005 - 16:45

It's nice to see this discussion come full circle. Getting "out of hand" is another way to put it, but I understand the passion that's gone into everyone's comments here.

It is indeed a commendable concept, but due to the nature of the beast (recording any kind of classical music), there are far too many variables, and I fear the end result is a kind of recording and methodology that many of us here may want no part of.

Frankly, I'm not interested in being told how to do my job at this level of minutia, have my integrity questioned, and being handcuffed to a limited approach that I can already see may have some serious conflicts to the task at hand. All that for a fixed fee, limited time to get the job done, etc., it sure doesn't sound like its worth the hassle. Fortunately, I have enough going on to keep me quite busy, thanks just the same, and I don't need the aggravation (which is why I'm mainly staying out of this whole discussion).

For example, I prepare demos and grant CDs several times a year for clients who are often competing directly with EACH OTHER for local grants, and unless there is a specific "no edit" clause in the competition rules, everyone edits, they'd be insane not to. Ditto for students and professionals on their way up the ladder. (Do we think classical music students are any less hip than other students their age? Editing is as common and accepted as surfing the web or downloading mp3's. )

Know that most serious organizations (for a very long time now) view demo/audition recordings as merely the start of the process, everyone assumes they've been edited and tweaked to put the artists' best foot forward. In addition, many candidates don't even get considered if they don't have a pedigree. A school like The Curtis School of music or an organization like the Philadelphia Orchestra gets hundreds of demos per year, many of which are rejected out of hand without the right credentials (letters of reference, teacher recommendations, existing track record, etc.)

The Academy of Vocal Arts (another merit-based tuition-free school like Curtis, here in Philadelphia that grooms tomorrow's opera stars) accepts audio recordings, but also requires a video as well, and always demands an in-person audition for final approval, occasionally even travelling OUT to a candidate if they deem the person worth the risk. The audio recording is merely the start of the process.

To sell this service to potential clients, implying that strict guidelines and "no edit clauses" will actually help their changes of getting work is, well, just plain wrong, IMHO. It just doesn't work that way out there in the cruel, cold, cutthroat world of professional musicians.

Arguing over "to edit or not to edit" is pointless. Everyone does, all the time. Consider this: Unless a project is undertaken to specifically NOT edit (which we all know is now a novelty, NEVER the norm), editing happens. All competing artists know that if they don't edit, the next one will.

The idea of forcing a starving, limited-funds struggling artist to accept or reject a half-hour (or whatever the arbitrary time limit) should things go horribly wrong seems just cruel, IMHO. (Ooh yeah, get that $$$ up front in case the kid vapor-locks, right?!?!?) I understand rules need to be in place to avoid wasting our time, but experience has shown me that 99% of the time, this is a self-policing situation all on its own. Someone gets sick, gets stuck in traffic, has a family emergency; anything can happen, and if I took a hard line like that with everyone that it ever happened to, I'd be out of work in no time.

I want no part of being some kind of timekeeper/traffic cop when 5 more minutes or one small edit might fix the whole thing and get the kid on his/her way to the next step in their career. This goes completely against my working methodology and audio "Bedside manner." Unless you've got icewater in your veins, NO ONE thrives or gives their best effort this way, at least at the early stages of a budding career. The method/rules being discussed here seem to turn the recording process itself into a contest as well. I simply don't work that way, and I suspect few others do either.

I do applaud & commend any efforts to standardize some kind of audition recording process, but I think it's going to have to be a basic set of rules, and go with that. Any more, and you'll never be done, and no one will be happy.

Besides, all the editing in the world won't let someone KEEP the job should they get it. If they can't cut it live, it'll show up soon enough. True artistic ability, while rare in other genres is a big part of what makes classical music tick. You either have it or you don't.

I'd be interested in a general system of local engineers making reasonably priced recordings for savvy artists looking for a good deal, but I'm definitely NOT interested in bioemetrics (you're KIDDING, right?!?!?), mic lists, "no-edit" clauses and god knows what other silly rules still to come - all for a fixed fee.

Thanks, really, but no.

x

User login