Audio source files

There's a lot to digest here. This FAQ topic is divided into four separate sections: 

Section A pertains to audio source files created here in my studio. 
Section B
pertains to audio source files which are brought, or sent in, from the outside (such as files originally recorded at another studio, at home, or downloaded from the internet). 
Section C
pertains to "stems" which are clusters - or submixes - of the individual tracks that make up a song project.
Section D pertains to the transmission or transportation of source files (i.e. how do we send files back and forth?)

The term "source files" refers to the individual audio clips that make up a digital multi-track recording project. A typical rock song (these days) can have anywhere from 8 to 100+ individual tracks of audio that have to be mixed together to form the completed song. Each instrument or sound you record becomes an audio file on my hard drive.

A. Source files created here.

The most common question related to source files is "Can I get the source files for my recording project to take with me?". Yes! And you should probably consider doing that out of principle no matter where you record. Some artists like to keep the original tracks, or source files, in the event they should ever want to go to another studio for additional recording or editing. Many musicians record parts here (drum tracks for example) and then take the files home to add the finishing touches. Whatever the reason, they belong to you so don't be afraid to ask for them. 

* The files that I create here are in standard WAV format and unless you tell me differently, they will be provided to you at the same bit depth and sampling rate as originally recorded (24 bit, 48 kHz for most projects). I can down-convert the WAV files to 16 bit/44kHz if you prefer. I can also provide the source files to you in Apple AIFF format or MP3. 

* Each WAV file is named or identified based on the instrument or track name (kick drum, snare drum, rhythm guitar, bass guitar, piano, lead vocal, etc.). If you are going to be using these files at another studio, I will create some track/session notes and contact information in the form of a readme file that I will include with the bundle. This way, the receiving engineer will at least have a reference point to start from, and a way to contact me if there are any problems or questions.

* Files which have been exported out of my recording program all start at what I call "time zero". Time zero is just an arbitrary starting point for the song. A song project might start out with the drummer clicking his sticks to establish a tempo for the band. Or maybe someone said "go" and the band just started playing. Who knows. The point is, I establish a time zero starting point a few seconds prior to the start of the performance. I also establish an ending point at the very end of the performance (usually after all the instruments have stopped and the audio has faded to dead silence). These points in time become time markers in my recording program. Once the starting and ending marker positions are established, they remain permanently fixed at those points in time so I can go back at a later date and export the audio from the exact same positions in time. If there are 16 tracks of audio then I would be creating 16 individual WAV files, all starting at time zero. You would import all those WAV files into your computer or recording device and make sure all the files start at the same exact time and everything will line up. This concept of time zero means there will be nothing but silence in some of the WAV files at the start and ending of each file. A lead guitar part might not start until 2 minutes into the song. Prior to that, there is dead silence. It's up to you whether you choose to strip away those silent portions of each WAV file in your audio editor.

* I should mention that there are other export/delivery alternatives such as OMF, broadcast WAV, bundled project files, etc. These alternate formats have merit and I can provide the source files to you in one of these if you prefer. If you don't know what any of that means, and you think it might be important, call.

* Unless otherwise specified you are getting the raw, dry, unprocessed WAV files (exception: see the section on Stems below). When I record a band or an artist here, the signal from each microphone or instrument ends up going directly to the computer hard drive. The analog signals are converted to digital, and then sent to the computer and stored on the hard drive as individual WAV files. These raw WAV files don't include all the mixing, panning, EQ settings or effects that I would normally apply to the final stereo mixdown. I generally don't record with outboard effects or any other external processing except maybe a bit of compression or some slight EQ adjustments if needed. There are plenty of exceptions though. A guitarist might choose to record a guitar track with some effects pedals. Tracks recorded with effects at the source will obviously still have those effects on the recorded track. My point is, if you intend to take the source files to another studio for additional recording or editing, you will have to create/recreate all those "mix" settings from scratch (which is usually the whole point of doing that in the first place). It's not uncommon for an artist to do the tracking in one place, mixing in another and mastering in yet another location. Happens all the time in the big leagues. 

* I archive all the source files for all the projects that I record here. If a client wishes to return for additional recording, all I have to do is load the project files from the original archive and away we go. 

* The question of who actually owns the source files can be an issue sometimes. It has been an issue here a few times already. It can get ugly. I've seen cases where bands break up literally during, or right after, the recording session and the issue of who has the legal right of ownership of the songs/source files comes into question. If you're working with a producer or manager, it can get even more complicated. Ownership of the source files is important because whoever has possession of the source files has the ability to go back and change or edit those files at a later date. Sometimes I get caught in the middle. Band members will often call me weeks or months or years after the project was finished and want me to provide them with the source files. I have no real way of knowing who actually owns, or has access rights, to the source files. My default position is that the owner is the person who paid me for the session - even if that person was not an actual band member. If you believe this will ever become an issue, I suggest you discuss it within the band and decide this before you come here (or anywhere) to record your songs. I will not be held accountable for any legal issues or gastro-intestinal disorders arising out of handing over source files to the wrong person. If someone other than the person who paid me wants the source files, I will attempt to make good faith contact with the band leader or person who paid me and confirm ownership. All it takes is a disconnected cell phone or non-functioning email address though and all bets are off.

B. Source files created elsewhere.

Clients often bring in audio files for mixing/mastering or additional recording and editing. Sometimes these files were recorded at other studios or perhaps at home. Sometimes the files were downloaded off the internet. Here are some general comments and observations to consider if you are bringing audio files from other locations.

* I work primarily with WAV files which were recorded and saved in the following formats: 16 or 24 bit, 44.1kHz or 48kHz

* MP3  is also an acceptable format for audio files, and they can be used as part of a multi-track recording project. I get a lot of karaoke singers who bring backing tracks in as MP3 files. Or sometimes sound effects or other elements are added to song projects in the form of MP3 files.

* Make sure the multi-track WAV files you are providing to me all start at what I call "time zero". What this generally means is that the engineer has exported (or "bounced") each track of audio to a new file with a common starting point. Each exported file contains a certain amount of dead space or quiet time in the beginning so that when I import the files into my recording program, all the positions of all the audio bits will line up properly. 

* Usually, in order to export/bounce tracks of audio for transporting, you have to be careful and make certain adjustments in the software to export the tracks correctly. Different programs handle this differently sometimes. I would not take this for granted. I would spot check the files after you create them by trying to re-import them back into the recording program and actually see with your own eyes (and ears) that the data has been exported correctly. I've gotten lots of files in the past that were "empty" or were dead silent from start to finish. If you re-import them back into the recording program you should at very least see something resembling an audio waveform. If there's nothing there, then obviously there was a problem during the export/bounce routine. Another quick way of verifying this is to look at the size of the exported files (using Windows Explorer). If the files all are the same size, that's a good indicator that the files were exported correctly (because they all start and end at the same time). If the files are all sorts of weird random sizes, then that's a good indicator they were not exported correctly and will not line up properly when imported into another system. Be careful about this though - Broadcast WAV files can be different sizes because there is time signature data encoded into the WAV file. I'm always suspicious of Broadcast WAV files but if that's all you got and they import properly into my recording app then we're golden.

* Please make sure that the engineer or person who creates the data discs provides some meaningful descriptive text which accompanies the project. I prefer it when the data files are given meaningful names like "kickdrum.wav" or "snare.wav", etc. Too often I receive files ambiguously labeled "track1.wav", "track2.wav", etc. which means that I must now sort through each file and try to make heads or tails out of the mess before I can do any meaningful work. A readme file with helpful background information is welcome. Contact information for the original recording engineer is even more welcome in case of questions.

* If the original audio was recorded as a mono file, make sure the exported file is also mono. If it was recorded in stereo, then the exported file should also be stereo. There is no point in exporting everything as stereo files. It just makes the entire project more difficult to transport because the files are twice as large as need be. When I see nothing but stereo files, that's a good indicator that the engineer wasn't paying attention when he created the transportable files (and there will usually be other problems as well). It's ok to export stereo files as individual left & right mono files but you don't need to do that. A single stereo file is fine.

* If the original files were recorded at 24 bit (or 32 float) then they should stay at 24/32 bit. Don't dither the files down to 16 bit unless you have to for some reason I can't think of right now. If they were recorded at 16 bit then leave them at 16 bit. Don't upconvert them. 

* One big caveat when exporting files from various recording programs: without trying to get too technical here, be careful about the volume level of the exported audio. Some programs use special "pan laws" which change the overall volume of the exported tracks. Also make sure you're exporting the raw audio and not exporting audio with pan or eq or effects. This gets tricky and it's hard to explain. As mentioned before, I would re-import the audio back into the program and confirm that the exported files match the source files in terms of volume levels. If you don't do this, there is a risk that the exported audio might be damaged due to overloads (exceeding digital zero decibels).

* I should also mention portastudios. I get a lot of projects here that were originally recorded at home on a portastudio. The word "portastudio" refers to tabletop multi-track recording machines that have a built-in hard drive, mixer, effects processor, etc. It's basically a studio-in-a-box that you can pick up and carry around. Retrieving audio files from these things are always a fun exercise. It seems no two machines handle audio storage and  retrieval  the same way (even from the same manufacturer). In some cases, you need a special software utility program running on the receiving computer to retrieve the files. It's up to you to make sure you have whatever it is you need to retrieve the files correctly. I would advise you not to wait until you get here and hope for the best. Do your homework ahead of time and make sure you have the tools needed to get the job done. If you're not sure, contact the manufacturer of the portastudio and find out what you need. I'm not an expert on every portastudio ever made. Don't cop an attitude with me if I don't understand exactly what buttons to push to get the audio out of your machine. It's your machine. You figure it out before you get here. Bring the user manual along if you have one just in case.

THE BOTTOM LINE. I've seen sessions ruined because the files brought or sent to me were damaged or improperly formatted. You can't be too careful about the integrity of the source files. Don't trust anyone but yourself. If you don't trust yourself to confirm the validity or integrity of your source files, seek out someone else to help verify things before you go running off half-cocked. It's frustrating and disappointing to schedule studio time and potentially have to drive for hours only to find out what you are bringing is useless. If you're coming from another studio, make sure they don't delete the original files before you've had a chance to confirm everything is OK.

If you're interested, here is a link to an ad hoc set of guidelines developed by the Producers & Engineers Wing of the Grammy organization (Recording Academy) which goes into a lot of extra detail about all this. Worth reading if you work in this business.

C. Stems

The word "stem" refers to groups or sub-mixes of tracks that make up a final song project. Stems are generally required for some sort of television or movie production work (also video game consoles).

Consider this: a typical rock/pop song could include as many as 10 tracks of drums or more (kick, snare, toms, cymbals, etc), a bunch of guitar tracks, a bunch of vocal tracks (including the lead and backing vocals), keyboards, percussion, etc. A television producer or director might require your multi-track song arrangement to be provided in the form of pre-mixed or pre-configured stems so they can edit it for a specific scene.

Here's a typical stem scenario:

Stem 1: kick drum 
Stem 2: snare drum 
Stem 3: all remaining drums in stereo
Stem 4: overheads and cymbals in stereo
Stem 5: bass guitar 
Stem 6: all guitars in stereo
Stem 7: lead vocal 
Stem 8: all backing vocals in stereo

The idea here is that important elements such as lead vocals, kick, snare, bass guitar, etc are treated as individual files which can be easily adjusted in relation to other elements. Typically all the tracks are rendered with whatever effects and other processing were used for the final master but that depends.

What we're really doing here is simplifying the mixing and editing process for the receiving engineer. The assumption is that the entire song had been mixed and mastered already so all that's really needed is some simple editing and mixing work to get the songs ready for the final project.

The people that you're sending stems to will usually specify how they want these things "stemmed" and what format they want the audio files in (24 bit vs 16 bit for example). I simply give 'em what they ask for and my job is done. 

Here's an example link to the stem information/instructions related to the Rock Band gaming system:


D. Delivery (transmitting/sending audio files)

The bane of my existence (lol). Transmitting (or receiving) audio files/projects from here to there. What's the best, safest and most efficient method of delivery? The answer is: it depends.

The basic problem is the size of the individual files. WAV files are large enough to create a headache when it comes to getting them from point A to point B. Even MP3s in large enough quantity can be an issue some times. Consider this: a four minute mono kick drum track/WAV file recorded at 24/48 will end up being approximately 40MB in size. Multiply that by the number of tracks and the number of songs and it adds up quickly. A single song's worth of WAV files can be anywhere from 1-2GB in size. 

Here are the various and more popular options for getting project files files back and forth.

Option 1 - snail mail the files on CD-R/DVD-R discs. Put all the files on standard CD-R or DVD-R data discs and mail them. Not a bad solution for local delivery especially for large amounts of data. Maybe not the fastest method but one of the safer methods because you get a hard copy of the files. If you can afford to wait I'll just send them USPS first class. For local delivery it's only a day or two. I usually eat the cost for first class ground delivery anywhere in the US. For out of state (or country) UPS or FedEx might be better options. You eat the cost.

I use specially designed cardboard shipping mailers. I've never had a problem shipping discs sent this way. You can usually find suitable shipping containers at any decent office supply store. I think the post office might also carry a small assortment of packaging materials. If you are living in a cardboard box under the tollway then this method might not work very well.

Option 2 - internet distribution. Obviously quicker and more efficient than snail mail but electronic distribution of large sums of data can get tricky depending on how much data we're talking about. If I'm the sender, and we're talking about less than 2 gigabytes worth of data, I can simply upload the files to my studio file server. I'll just create a secure user account with a password and you can download them directly to your computer using Internet Explorer or Firefox or whatever browser you normally use. 

Option 3 - Hightail or Dropbox or some other online file exchange service. Works well but depending on the amount of data, you would probably have to pay a monthly fee. Clients have sent me large amounts of data through these services and they tend to work pretty well. Dropbox has a "pro" plan which costs about $10 a month and allows for around 100GB of file storage - enough for an entire album of data files. If you have a paid account then you would have to provide me with your username and password if you want me to upload to your account. Downloading is handled automatically (I get an email telling me files are ready to be downloaded and a link to the service directory - it's pretty simple).

Option 4 - Email. works ok for a couple of MP3 files. I can attach a file (or files) to a simple email message. My email account limits the total size of attachments per message (incoming and outgoing) to 10 MB. 

ZIP - the ZIP data compression utility is used to bundle a group of files for electronic delivery. I would normally create a large ZIP file for each song title. You unZIP them on your end. The ZIP utility is included with MS Windows.


Back to FAQ