Music in Audio Documentary at DVinfo.net

Jason McGovern · February 16th, 2009, 09:58 PM

Hello, can anyone give me any advice on using music in an audio documentary - kind of the way they do it on "This American Life"? When you're mixing music into a piece as a way of bridging two vocal tracks, what is a good approach to take? What kind of levels should one aim for? Any help/links much appreciated. I'm hoping to submit this to my local Public Radio program.

Also -- I have both Premiere and Audition. I was thinking about just organizing it in Premiere - seems like it's file browser is a lot more robust than Audition. Plus I can have my source file open above the actual Program tracks. In Audition you constantly have to switch back and forth between Edit and Multitrack.

Michael Nistler · February 17th, 2009, 10:51 AM

Hi Jason,

I don't recall watching "This American Life" but enjoy lots of documentaries with music mixing between vocal tracks.

Depending on your workflow and what you're trying to accomplish, you can certainly use either Audition or Premiere. Of course, Audition would be the way to go if you are doing a sophisticated mix with lots of tracks, effects, etc. But for your basic documentary mixing needs, it's easy enough to add channel tracks to Premiere, add some audio keyframes and fade in/out the music and vocal tracks.

Good luck, Michael

Seth Bloombaum · February 17th, 2009, 11:22 AM

Quote:

Originally Posted by Jason McGovern

Hello, can anyone give me any advice on using music in an audio documentary - kind of the way they do it on "This American Life"? When you're mixing music into a piece as a way of bridging two vocal tracks, what is a good approach to take? What kind of levels should one aim for? Any help/links much appreciated. I'm hoping to submit this to my local Public Radio program...

Usually we have to avoid commercial music recordings due to the difficulty and expense of securing use licenses from copyright holders (in the USA). My understanding is that with the Public Radio distribution you have in mind use of such music would come under the annual ASCAP/BMI licenses that broadcast stations buy for use in programming - so, your distribution would likely be fully licensed. Maybe you knew that, but commercial music is a touchy subject and the broadcasters have a special position for their programming.

When you have this music, from CD or whatever, and you're mixing, it is usually very important to bring the music down *at least* 6db. Most is very hot, lots of production time went into making it as hot as possible. You need good monitoring and good ears to properly mix and balance this with your interview and narrative trax.

You'll need to consider compression and/or loudness maximizing for your vox trax. Then, you'll quickly notice any unwanted noise in the vox, as either of these techniques will tend to bring up background noise.

Without these steps commercial music will tend to overpower your vocal tracks. Notice that on This American Life they frequently use music with very spare arrangements. Not a lot of energy - not distracting from the intended content.

And do visit transom.org for more advice, really a treasure-trove, from people who are producing This American Life and other radio storytelling.

Jason McGovern · February 20th, 2009, 07:26 PM

Thanks very much! Excellent advice. From what I'm able to determine - a lot of times This Life will begin an audio track well before the person is done speaking - and you're right it is very sparse - and then when the person is done - they pull the levels up - but from what is mentioned above - even then they still don't go up to 0db I think.

I'm laying down the narration parts now with a crappy Sure mic and a Marantz PMD660 and it doesn't sound bad. The hard part is trying to decide how much inflection to use. I've noticed on This Life - the host Ira comes across very informally while still retaining a marvelous sense of clarity - but a lot of the reports sound so dry - is it me or do they all start to sound the same? Which makes me think that that is an aesthetic choice?

Oh well - thanks for the advice.
J

Ty Ford · February 21st, 2009, 09:57 AM

The average public radio performance is pretty dry. Then there's Klick and Klack and Garrison Keillor from Prarie Home Companion, both of which show a lot more spirit.

If you know someone at NPR, you might ask them for direction.

Regards,

Ty Ford

John Harrison · February 23rd, 2009, 01:49 PM

I haven't worked in broadcast Radio but for audio levels in Broadcast tv by the time its ready to air, the headroom is usually vox running at around -14db with music going NO higher than -6db (VU) , I would imagine radio is the same and US and European broadcast standards are similiar... i

Jason McGovern · February 23rd, 2009, 06:26 PM

Hey John, can you explain what that means?

As an experiment I downloaded a This American Life episode from their favorites list. It's called Act V and it's marvelous, but buying it through ITUNES the bitrate is obviously not good. So I recorded a sample where I really like how they fade the music in underneath the voice - then bridge - then fade down into the next voice.

Next I normalized the section to -1 DB and went in and took a look at db for music underneath voice - is this called bed? hmm... I looked at the snare on a hip hop sample since this was the highest peak - the loudest snare on the bed coming in at -8db and the same snare on the bridge at -1.7 DB. It was interesting cause when i try to do my own by ear I always feel like the audio is too loud - but looking at this as a sample - it seems like they keep the bed pretty loud compared to what I imagined it would be.

John i'm not sure what -14db vox means...

I sure do wish they had the same Audio Effects in Premiere as in Audition - well at least the presets anyway. For the clipping issues it looks like I'm just going to shear off the tops using the Hard Limiting in Audition then import them into Premiere. Using the Source clip window and the Timeline just makes the workflow so much faster than in Audition. If I was doing a lot of Sends with different effects then I would use Audtion - I took some Lynda training on Audition and the stuff you can do is amazing. but for my purposes, I'm going to stick with Premiere.

As far as finding the actual music - it seems like the most utilized bridge/bed music on TAL is music featuring nice hip hop beats - and if it's an emotional section - a simple piano part thrown in. They use a ton of different music, but from what I've heard - hip hop is the cornerstone.

I've found ccMixter - Welcome to ccMixter ! Amazing. Creative commons hip hop beats galore. Plus in a lot of instances - you can download the actual tracks and mix them yourself. I think I might do that with some garageband just to drag this project on as long as I can ;)

Jason McGovern · February 23rd, 2009, 06:48 PM

QUESTION: It looks as though every time i either export from Prem or Mixdown from Aud ... it looks like both programs are doing some kind of normalizing/limiting on it's own. I tried one file with a lot of clipping - made no edits and did straight export/mixdown - and with both programs i'm seeing something going on... some kind of limiting. what gives?

John Harrison · February 26th, 2009, 11:55 AM

Hi Jason

Sorry if i wasn't a bit clearer, you were asking about levels (and levels and loudness are completely different). Usually what your listening to (or was when i was a tx editor-TV) is voice(vox) broadcast at around the -14 db range with music peaking at -6db.This is to stop
peaking in transmission

So when you set 0db tone on the front of your recording this is aligned by the broadcast engineer to (scratches head and trys to remember) -18db, Naturally there are compressors to prevent peaking anyway but those are the rough guidelines (from tv) but any station should provide technical specs acceptable on anything recieved including levels and formats

... I half answered to find out if radio was the same......

Hope thats clearer (if not much help)

February 16th, 2009, 09:58 PM	#1
Jason McGovern Regular Crew Join Date: Mar 2008 Location: Chicago, IL Posts: 35	Music in Audio Documentary Hello, can anyone give me any advice on using music in an audio documentary - kind of the way they do it on "This American Life"? When you're mixing music into a piece as a way of bridging two vocal tracks, what is a good approach to take? What kind of levels should one aim for? Any help/links much appreciated. I'm hoping to submit this to my local Public Radio program. Also -- I have both Premiere and Audition. I was thinking about just organizing it in Premiere - seems like it's file browser is a lot more robust than Audition. Plus I can have my source file open above the actual Program tracks. In Audition you constantly have to switch back and forth between Edit and Multitrack.

February 21st, 2009, 09:57 AM	#5
Ty Ford Inner Circle Join Date: Nov 2004 Location: Baltimore, MD USA Posts: 2,337	The average public radio performance is pretty dry. Then there's Klick and Klack and Garrison Keillor from Prarie Home Companion, both of which show a lot more spirit. If you know someone at NPR, you might ask them for direction. Regards, Ty Ford __________________ Ty Ford's Audio Bootcamp Field Guide Ty Ford's Audio Video Blog

February 17th, 2009, 10:51 AM	#2
Michael Nistler Major Player Join Date: Aug 2006 Location: Petaluma, CA Posts: 456	Hi Jason, I don't recall watching "This American Life" but enjoy lots of documentaries with music mixing between vocal tracks. Depending on your workflow and what you're trying to accomplish, you can certainly use either Audition or Premiere. Of course, Audition would be the way to go if you are doing a sophisticated mix with lots of tracks, effects, etc. But for your basic documentary mixing needs, it's easy enough to add channel tracks to Premiere, add some audio keyframes and fade in/out the music and vocal tracks. Good luck, Michael

February 20th, 2009, 07:26 PM	#4
Jason McGovern Regular Crew Join Date: Mar 2008 Location: Chicago, IL Posts: 35	Thanks very much! Excellent advice. From what I'm able to determine - a lot of times This Life will begin an audio track well before the person is done speaking - and you're right it is very sparse - and then when the person is done - they pull the levels up - but from what is mentioned above - even then they still don't go up to 0db I think. I'm laying down the narration parts now with a crappy Sure mic and a Marantz PMD660 and it doesn't sound bad. The hard part is trying to decide how much inflection to use. I've noticed on This Life - the host Ira comes across very informally while still retaining a marvelous sense of clarity - but a lot of the reports sound so dry - is it me or do they all start to sound the same? Which makes me think that that is an aesthetic choice? Oh well - thanks for the advice. J

February 23rd, 2009, 01:49 PM	#6
John Harrison Regular Crew Join Date: Feb 2009 Location: Dublin Ireland Posts: 34	I haven't worked in broadcast Radio but for audio levels in Broadcast tv by the time its ready to air, the headroom is usually vox running at around -14db with music going NO higher than -6db (VU) , I would imagine radio is the same and US and European broadcast standards are similiar... i

February 23rd, 2009, 06:26 PM	#7
Jason McGovern Regular Crew Join Date: Mar 2008 Location: Chicago, IL Posts: 35	Hey John, can you explain what that means? As an experiment I downloaded a This American Life episode from their favorites list. It's called Act V and it's marvelous, but buying it through ITUNES the bitrate is obviously not good. So I recorded a sample where I really like how they fade the music in underneath the voice - then bridge - then fade down into the next voice. Next I normalized the section to -1 DB and went in and took a look at db for music underneath voice - is this called bed? hmm... I looked at the snare on a hip hop sample since this was the highest peak - the loudest snare on the bed coming in at -8db and the same snare on the bridge at -1.7 DB. It was interesting cause when i try to do my own by ear I always feel like the audio is too loud - but looking at this as a sample - it seems like they keep the bed pretty loud compared to what I imagined it would be. John i'm not sure what -14db vox means... I sure do wish they had the same Audio Effects in Premiere as in Audition - well at least the presets anyway. For the clipping issues it looks like I'm just going to shear off the tops using the Hard Limiting in Audition then import them into Premiere. Using the Source clip window and the Timeline just makes the workflow so much faster than in Audition. If I was doing a lot of Sends with different effects then I would use Audtion - I took some Lynda training on Audition and the stuff you can do is amazing. but for my purposes, I'm going to stick with Premiere. As far as finding the actual music - it seems like the most utilized bridge/bed music on TAL is music featuring nice hip hop beats - and if it's an emotional section - a simple piano part thrown in. They use a ton of different music, but from what I've heard - hip hop is the cornerstone. I've found ccMixter - Welcome to ccMixter ! Amazing. Creative commons hip hop beats galore. Plus in a lot of instances - you can download the actual tracks and mix them yourself. I think I might do that with some garageband just to drag this project on as long as I can ;)

February 23rd, 2009, 06:48 PM	#8
Jason McGovern Regular Crew Join Date: Mar 2008 Location: Chicago, IL Posts: 35	QUESTION: It looks as though every time i either export from Prem or Mixdown from Aud ... it looks like both programs are doing some kind of normalizing/limiting on it's own. I tried one file with a lot of clipping - made no edits and did straight export/mixdown - and with both programs i'm seeing something going on... some kind of limiting. what gives?

February 26th, 2009, 11:55 AM	#9
John Harrison Regular Crew Join Date: Feb 2009 Location: Dublin Ireland Posts: 34	Hi Jason Sorry if i wasn't a bit clearer, you were asking about levels (and levels and loudness are completely different). Usually what your listening to (or was when i was a tx editor-TV) is voice(vox) broadcast at around the -14 db range with music peaking at -6db.This is to stop peaking in transmission So when you set 0db tone on the front of your recording this is aligned by the broadcast engineer to (scratches head and trys to remember) -18db, Naturally there are compressors to prevent peaking anyway but those are the rough guidelines (from tv) but any station should provide technical specs acceptable on anything recieved including levels and formats ... I half answered to find out if radio was the same...... Hope thats clearer (if not much help)