Hello,

I was commissioned to write theme music for a podcast. I finished the track and sent it their way for review, but after listening to some of their previous episodes I noticed that the music they were using before (various excerpts of songs they like) sounds extremely compressed. I know that podcasters will often use heavy compression so that their voices don't jump in levels, which is fine, but when this compression is happening during the intro and outro music, it really makes the music sound bad. On top of this, I'm sure podcast episode file is a super compressed MP3 or something similar which also hurts the overall quality.

I guess my question is do you know of a way to ensure the highest quality possible once my music is implemented into their podcast? Is there anything I can do on my end (regarding file type, etc.). Also, is there anything that they should be doing on their end when they are compressing their voice audio that I could suggest to them? Any ideas are welcome. Thank you!

Once you give it to someone else, it really is out of your control to what it becomes.  I have heard some great recordings botched due to improper handling by podcasters.  The best thing you can do is bring it up to the person and suggest a different procedure to ensure better quality.  Most likely they are using something like audacity with little quality control, know-how, and very low bit rate mp3 output.  

yes,
most folks that do pod casts,
don't have a clue what they are doing tech wise.

I was just going to make bats' point.  It is like everyone posts a blog w/o having any idea of the craft of writing.  Thank the gawds for spell check or you'll think you are reading a foreign language.
 
Same with the quality of sound.  most podders know what a good one sounds like, but have no idea how to accomplish it.  And that isn't the point of a pod cast, or their reason for doing one.
 
@

The greatest irony is the number of lo-fi podcasts that are dedicated to audio-related topics.
 
I've always written it off to low bit rates. Many of them are still at 64  kb/s or 96 kb/s. Some as low as 32 kb/s (!). 

Any suggestions on how to tactfully ask them what their process is? If they're using low bit rates or something, it could be as simple as just telling them to change their settings. What about compression? Would it be a good idea to process the speaking portion of their podcast separately so that they are only compressing their voices? Then they can add the intro and outro music as separate files to the beginning and end of the podcast, combining it into one file using Audacity or whatever it is they're using? The only problem with that is that they asked me to write the theme in a way that they can speak over the last 15 yo 30 seconds or so. So, the main theme is the first 20 seconds, then there's a thinned out arrangement that they would speak over. I'm trying to figure out how they could handle this. Maybe just start their voice compression (if they're actually using any) from that point? With the music turned down it probably won't effect it too much. 

Sorry, I'm kind of thinking "out loud" here. :)

A podcast itself has no limitations on it's quality... You can post a 32 kb/s or a 384 kb/s.... it's up to the person doing the final production and preparation.
 
1: They may not know what they are doing... and blindly uploading horrible quality. You can help them with that, if that is the case.
 
2: Podcasts are served from "your server" so file size may be critical. Some may keep the quality low to avoid bandwidth costs if it should be really popular. Or, if someone is serving from home keeping the file size really small may be the only option due to horrible upload speeds from most providers.

File size is important to podcasters, because unlike music downloads, podcasts tend to be at least 20 minutes in duration and many are over an hour in length.
 
There is a podcast that I enjoy called The Rock and Roll Geek Show. Most shows are 60-80% music and they go out at 256 kb/s. Normally over an hour in length, typical file sizes are around 120-150 MB. That's an acceptable download for those who are already fans, but it doesn't encourage new listeners to take a chance on it.
 
At the other end of the quality spectrum is another of my favorites, the Sonic State podcast. Those episodes are also fairly long and feature a lot of music, but they're released at 96 kb/s or low-quality VBR averaging 80 kb/s. File sizes are 30-50 MB. I cringe whenever they play music.
 
One other observation: it takes forever to edit a podcast well. It is mind-numbing drudgery. I can completely understand why so many accept a lower standard of quality. Anything you suggest to them to improve music quality will have to be simple, fast and easy.
 
I'd first try the two things that are easiest: use VBR and separate compression for music and speech. The former is just a command line option in LAME, the latter merely requires the addition of a bus to route the spoken tracks to for compression.
 
The tricky part is going to be finding the right balance between music and speech compression. If the difference is too great you're going to have trouble getting the music loud enough. But once you've figured it out they'll simply set up their template with the music compression baked in and the voice compressor already set up.
 

When I encounter something like that and I have to bring it up with a client, I simply start asking questions.  What bit rate are you using? Are you sure you exported a stereo file? (you may know they didn't) Etc, etc.  This allows the other person to come to their own conclusion that they might need help or at least pay closer attention.  There is a balance between still wanting to work and not having your work compromised.  Gain trust, and don't bruise their ego, so they feel they can rely on your expertise.  Most people I know are open to at least a small nudge in the right direction if it leads to better results.  

Thanks for all the responses and insights, everyone. Quick update. After a chat with the podcasters, I learned that they had been ripping the music they had been using from YouTube and using those MP3 files. So, we already know where the problem started. I think just having a high-quality file to start with will make a big difference. I asked them to let me hear the episode that has my intro music before they publish it so I can check the quality and see if there's anything more I can do to help them keep it high. :)