Compressing sermon mp3s for podcasting

[David Pettigrew]

I recorded my sermon in mp3 format yesterday, and announced it would be available for podcast. My problem is it created about a 30mb file, and I'd really like to get that down under 8mb, the limit for uploading to our church website (wordpress software.)

Any suggestions for a good compression program? I'm willing to pay for one if it really works well without reducing quality too much.

[Jim Abrams]

Preach shorter sermons and use simple words that need less explanation :-)

[G R 'Scott' Cundiff]

I recorded my sermon in mp3 format yesterday, and announced it would be available for podcast. My problem is it created about a 30mb file, and I'd really like to get that down under 8mb, the limit for uploading to our church website (wordpress software.)

Any suggestions for a good compression program? I'm willing to pay for one if it really works well without reducing quality too much.


First, for future recordings record in 16bit, 32000 mono. It'll be good enough for the spoken voice and will leave you with a file around 10 megs in size for a 30 minute sermon.

To convert it down, get Audacity and Lame plugin. If you read the docs, you'll find how to resample what you have now to the settings I named above.

It's a freeby:
http://audacity.sourceforge.net/
http://audacity.sourceforge.net/help/faq?s=install&i=lame-mp3

[Jon Twitchell]

First, for future recordings record in 16bit, 32000 mono. It'll be good enough for the spoken voice and will leave you with a file around 10 megs in size for a 30 minute sermon.

To convert it down, get Audacity and Lame plugin. If you read the docs, you'll find how to resample what you have now to the settings I named above.

It's a freeby:
http://audacity.sourceforge.net/
http://audacity.sourceforge.net/help/faq?s=install&i=lame-mp3

I had read that the LAME encoder is not as efficient as the iTunes encoder... I haven't done any testing, but I use the iTunes encoder. The big thing is to make sure that your final file is in mono... it'll cut your file size about in half. You can change all the settings in iTunes to get the file quality/size you want. I have one set of settings I use for ripping CDs, and another I use for encoding sermons.

[Billie Goodson]

I had read that the LAME encoder is not as efficient as the iTunes encoder... I haven't done any testing, but I use the iTunes encoder. The big thing is to make sure that your final file is in mono... it'll cut your file size about in half. You can change all the settings in iTunes to get the file quality/size you want. I have one set of settings I use for ripping CDs, and another I use for encoding sermons.

There is a difference between monotone and monotonous, right? :)

[Jon Twitchell]

There is a difference between monotone and monotonous, right? :)

Hmmm... not sure that there is, actually.

but "mono" is actually short for "monaural" or "monophonic." :)

[Jim Chabot]

I had read that the LAME encoder is not as efficient as the iTunes encoder... I haven't done any testing, but I use the iTunes encoder. The big thing is to make sure that your final file is in mono... it'll cut your file size about in half. You can change all the settings in iTunes to get the file quality/size you want. I have one set of settings I use for ripping CDs, and another I use for encoding sermons.

Ditto. I have found that the itunes encoder works great, it is free and easy to use. I still use audacity to edit the initial .wav file before encoding.

[David Pettigrew]

I had to split it into two files this time. It ain't pretty, but hopefully I'll do better next week.

http://crossview.info/index.php/2010/08/30/sermon-podcast/ (http://crossview.info/index.php/2010/08/30/sermon-podcast/)

[David Pettigrew]

Jon Twitchell helped me upload the file directly to the server (yourchurchweb.net). I compressed it down to 14mb. I had recorded it directly to my mp3 player, which is why the audio is so tinny. I'm going to try recording through the church mixer directly to the laptop this week. Thanks everyone for the help!

[Jon Twitchell]

I'm glad we could help you out! :)

A couple of things:

1) Can I suggest a different file-naming nomenclature? At some point, I had to change my naming rules to YYYY.MM.DD, instead of (M)M.DD.YY. By including all four YYYYs, you distinguish if that number is a year or a month... and by putting the year first, the files will show up in alphanumeric order. I'd also suggest tagging the file with one or two descriptive words after the date... which will enable you (later on) to find a sermon that you are looking for. For instance, 2010.08.29_hosea2, would not only help me know when, but roughly what the sermon was.

2) As you setup the laptop/soundsystem combo, remember that it's easier to amplify a weak signal then to fix a distorted signal that was too loud to begin with. In other words... err on the side of recording it too quietly, recognizing that you might preach louder then you're testing it out. If the signal peaks/clips, there's not much you can do to fix it... but if it's not quite loud enough, you can fix that in post-production. Also, you might be able to use an auxiliary send on your board, allowing you to record JUST the pulpit mic.

3) I run a couple of audacity effects to help with the audio... First, I normalize (and remove DC Offset) the track. Then I amplify the track (not allowing peaking), and lastly, I run the compressor. All of that helps to even out the volume differentials. Scott or Jim might have some better ideas (being much more into audio than I am...), but that works for me.

Good luck!

[Gary Creely]

3) I run a couple of audacity effects to help with the audio... First, I normalize (and remove DC Offset) the track. Then I amplify the track (not allowing peaking), and lastly, I run the compressor. All of that helps to even out the volume differentials. Scott or Jim might have some better ideas (being much more into audio than I am...), but that works for me.


You should compress it first, then normalize it. The compression (if properly set) will knock down the biggest peaks, normalizing then brings the remaining peaks to the top of your head room. The result is the program material that is using the total dynamic range available.

[Jon Twitchell]

You should compress it first, then normalize it. The compression (if properly set) will knock down the biggest peaks, normalizing then brings the remaining peaks to the top of your head room. The result is the program material that is using the total dynamic range available.

My first normalization isn't for amplitude... but to remove DC Offset. (I use a battery powered MP3 recorder that doesn't center the audio). If I don't fix that first, then the other effects will just augment that problem. So I don't check the "amplitude" box when I run the Normalize effect... just the DC Offset box.

The compressor in Audacity has a checkbox that reads "make-up gain for 0db after compressing." I believe that does what you're suggesting... running the normalizer after the peaks have been knocked down.

Do you think that's right, or should I try something different?

Thanks for the input!

[Gary Creely]

My first normalization isn't for amplitude... but to remove DC Offset. (I use a battery powered MP3 recorder that doesn't center the audio). If I don't fix that first, then the other effects will just augment that problem. So I don't check the "amplitude" box when I run the Normalize effect... just the DC Offset box.

The compressor in Audacity has a checkbox that reads "make-up gain for 0db after compressing." I believe that does what you're suggesting... running the normalizer after the peaks have been knocked down.

Do you think that's right, or should I try something different?

Thanks for the input!

Sounds like the compressor is also normalizing. I looked at the program a little closer, the only possible suggestion I would make to that flow is to not normalize it the first time through (just deal with the DC offset), so that you are not normalizing it twice. It is handy that they include the normalizing option on the compressor.

as a side note audacity is a really nice audio editing program considering it's free!

[Jon Twitchell]

Sounds like the compressor is also normalizing. I looked at the program a little closer, the only possible suggestion I would make to that flow is to not normalize it the first time through (just deal with the DC offset), so that you are not normalizing it twice. It is handy that they include the normalizing option on the compressor.

as a side note audacity is a really nice audio editing program considering it's free!

I think (but I don't know), that in my first normalizing, ALL I'm doing is dealing with the offset... it has checkboxes both for DC Offset and for amplitude. I have the DC Offset box checked, and the amplitude box unchecked.

Audacity is great... it's got a few quirks from time to time... and it's not quite as clean as what I was using in the radio studio... but it's great for what I use it for.

[Gary Creely]

I think (but I don't know), that in my first normalizing, ALL I'm doing is dealing with the offset... it has checkboxes both for DC Offset and for amplitude. I have the DC Offset box checked, and the amplitude box unchecked.

Audacity is great... it's got a few quirks from time to time... and it's not quite as clean as what I was using in the radio studio... but it's great for what I use it for.

Got it, the way it was originally describe sounded backward. But since normalize actually equals DC offset, and compression is actually normalizing that is a good process. Considering the gear you use to record your podcast it turns out rather well.

[Gary Creely]

In a related note I have been looking for a good way to host audio somewhere, but have not found any great solutions. There are so many options for pictures and video, but plain audio tuff. Yahoo had a nice service for a while, but it seems they discontinued it.

[Jon Twitchell]

In a related note I have been looking for a good way to host audio somewhere, but have not found any great solutions. There are so many options for pictures and video, but plain audio tuff. Yahoo had a nice service for a while, but it seems they discontinued it.

what would you want such a service to include?

[Gary Creely]

The 3 things I would like are:

1. to host the audio files
2. ability to imbed a player
3. free

[G R 'Scott' Cundiff]

The 3 things I would like are:

1. to host the audio files
2. ability to imbed a player
3. free

I use dropbox which is all of that: http://nazareneblogs.org/miscellany/2010/05/24/an-easy-way-to-post-sermon-audio-online/

[Gary Creely]

I use dropbox which is all of that: http://nazareneblogs.org/miscellany/2010/05/24/an-easy-way-to-post-sermon-audio-online/

That is a great way! I have had drop box for years and didn't realize that functionality. One thing about your tutorial, I don't think you can get a link to a folder now, just a file. (or at least I can't figure it out)

[G R 'Scott' Cundiff]

That is a great way! I have had drop box for years and didn't realize that functionality. One thing about your tutorial, I don't think you can get a link to a folder now, just a file. (or at least I can't figure it out)

On your computer go to the dropbox folder. Left click on the "Public" folder, pick dropbox, pick get sharable link. Works like a champ!

[Gary Creely]

On your computer go to the dropbox folder. Left click on the "Public" folder, pick dropbox, pick get sharable link. Works like a champ!

Both on my mac and PC "copy public link" are not available on on folders, but it is on a file.

Then I saw this on the dropbox site
942

[G R 'Scott' Cundiff]

Both on my mac and PC "copy public link" are not available on on folders, but it is on a file.

Then I saw this on the dropbox site
942

I should have said "right click" rather than "left click." When I right click on my Dropbox Public Folder I see this:
943

[Gary Creely]

I should have said "right click" rather than "left click." When I right click on my Dropbox Public Folder I see this:
943

Here is what I get:

955


I am thinking they may have omitted that feature in a later version.

[G R 'Scott' Cundiff]

I think you may not be using the version I referenced in the blog. It's at http://forums.dropbox.com/topic.php?id=22231

[Gary Creely]

Works great- I had the current version, and didn't look closely enough the tutorial. Thanks for the tip, this is by far the best option I found for easy hosting.

[G R 'Scott' Cundiff]

Glad you got it going. You can then wrap it in an iframe and put it on your website. I agree that it's the easiest approach around.

[Roy Richardson]

Can I ask an Audacity newbie question here? We actually record the entire service in CD format for some shut-ins. I am trying to figure out how to sequentially lay out tracks in Audacity, instead of layering them. The CD recorder is setting a new track about every 5 minutes, so the sermon is in pieces, so to speak.

I cannot figure out how to stitch these together in Audacity. Any suggestions?

[Jon Twitchell]

In audacity there is a time shift tool... it looks like a double-ended arrow... that allows you to shift the tracks to the right or left.

I suspect you are looking for an automated feature... but that's how you would do it manually.

Is there any chance you could skip the CD recorder and record into audacity first?

[Roy Richardson]

I'll have to check with my tech guy. He got me this far.

[Rich Schmidt]

The CD recorder is setting a new track about every 5 minutes, so the sermon is in pieces, so to speak.

I wonder if that's an option you can switch off. I think it is on our CD recorder. We just have it record one long track, which then gets ripped onto the computer, edited in Audacity, and exported as an .mp3 for the website.

[G R 'Scott' Cundiff]

Seems you could record the whole service in audacity .. Then burn it to cd directly from audacity .. Then export it as mp3.

[Roy Richardson]

I wonder if that's an option you can switch off. I think it is on our CD recorder. We just have it record one long track, which then gets ripped onto the computer, edited in Audacity, and exported as an .mp3 for the website.
I'll check into that going forward. I'm trying to get some past ones taken care of

[Rich Schmidt]

Is there any chance you could skip the CD recorder and record into audacity first?


Seems you could record the whole service in audacity .. Then burn it to cd directly from audacity .. Then export it as mp3.

I don't know about Roy's setup, but in our case, it's much easier to record to CD first, since the CD recorder is already in the rack with all the other sound equipment. The laptop running ProPresenter is right next to it, but I'd rather not depend on one laptop to run ProPresenter, occasionally show video clips, AND record the service. And I don't really want to set up a 2nd computer in the limited space we have back there just to record the service... especially when we already have a CD recorder in the rack.

[Roy Richardson]

I don't know about Roy's setup, but in our case, it's much easier to record to CD first, since the CD recorder is already in the rack with all the other sound equipment. The laptop running ProPresenter is right next to it, but I'd rather not depend on one laptop to run ProPresenter, occasionally show video clips, AND record the service. And I don't really want to set up a 2nd computer in the limited space we have back there just to record the service... especially when we already have a CD recorder in the rack.

Ditto

[Gary Creely]

I wonder if that's an option you can switch off. I think it is on our CD recorder. We just have it record one long track, which then gets ripped onto the computer, edited in Audacity, and exported as an .mp3 for the website.

Most CD recorders have three different track creating options. The first is a fixed period of time such as what you are dealing with, the second is one that looks for low signal and creates a track when the signal passing through the CD recorder falls below an adjustable threshold, and the last option is manual. One of the most common CD recorders in churches is the Tascam RW900, and to mark another track you simply hit the record button again and it seamlessly creates a track at that spot.

For most ministries there is a great advantage to recording directly to a CD because it can immediately be reproduced. The obvious limitation is 70 minutes of recording time. We have sold some CompactFlash recording devices, but at that point you might as well just record straight to a computer.

[Roy Richardson]

Well, I have 2 podcasts up at www.hobartnaz.com. I'm still trying to figure out the website CMS system, but progress is being made.

Thanks for all your help.