Stop making crappy MP3s with Audacity and LAME! I tell you how to encode MP3s the way LAME was designed to, so you get a great audio quality and small file size. I get a little more technical and use a lot of numbers and abbreviations, so hold on, but I hope it's worth it.
Podcast: Play in new window | Download
Not for podcasts!
Since recording this episode, I discovered problems with podcasts using the below method. This is still the preferred method for music, but may produce incompatible podcast files. Instead, I recommend exporting to WAV and using iTunes to create you podcast MP3s.
Use the latest Audacity!
In case you don't already have it, use the latest Audacity version, which improves a lot of things over older versions, adds better compatibility, and also made it easier to change your encoder options. Download here.
What file-compression and “codecs”?
In The Audacity to Podcast™, episode 5, I shared about compressing your audio to get a better, more evened-out sound. This time, I'm talking about compressing your files to save space but still maintain the highest quality you can.
“Codec” is short from compression/decompression. It's the way a computer compresses an audio or video file to make it smaller and thus easier to transfer (or fit within limited hosting such as LibSyn). The two most-popular MP3 encoders are LAME and Faunhofer.
LAME and Fraunhofer are both the best MP3 encoders!
How can they both be the best? Because they're designed to do different things. LAME is free and open-source and it's what Audacity uses. Fraunhofer is licensed and is included with high-end software, but can be used free with iTunes. Cliff Ravenscraft once welcomed me on Podcast Answer Man #125 for us to discuss bitrates. But Cliff remains strongly opinionated against LAME since he prefers CBR, and I'll explain that later.
Bitrates: kbps and KB
Audio bitrates are measured in kilobits per second (kbps), which are different from kilobytes (KB). There are eight bits in one byte. The default MP3 encoding setting is at the rate of 128 kbps.
Stereo vs. mono
Choose carefully whether you want to release your podcast episodes as stereo or mono. Stereo takes up double the space of an equal-quality mono file, because it has twice the tracks/channels. Please do not split your cohosts into left and right in your audio, unless you're doing a special dramatization that requires positional audio. Even then, the best place for voices is right in the middle: mono or balanced in stereo.
In Audacity, encoding a mono file at a bitrate of 128 kbps means the second channel will get 128 kilobits per second to capture the details of its audio. But a stereo file at the same bitrate will split the 128 kbps between the left and right channels, thus equaling the quality of 64 kbps audio.
In iTunes, you get the same final quality regardless of choosing stereo or mono. If you select 128 kbps stereo, it gives both channels 64 kbps. If you chose 128 kbps on a mono file, it gives the single channel still 64 kbps.
Variable bitrate (VBR) versus constant bitrate (CBR)
When you select 128 kbps with constant bitrate (CBR), the MP3 encoder will use 128 kbps to digital “describe” the audio no matter what is happening. But if you switch to variable bitrate (VBR), which I suggest, then the encoder will work within a range to use more space when the audio is more complex or us less space when the audio is less complex.
LAME is the best VBR encoder. It was designed to encode music at a variable bitrate and it performs this better than any other popular encoder at the same quality (even Ogg Vorbis in most cases). Fraunhofer is the best CBR encoder. It was designed to encode audio at a constant bitrate and it performs this with better-sounding quality than LAME.
Quality and file size comparisons
In case my re-encoding of my sample audio distorted the comparison. Here are my original MP3s that you can download for comparison. Please note that when I recorded these and this episode, I had accidentally left my secondary mic on and near something that was generating noise, hence the hiss.
32kbps Stereo CBR (82 KB)
Podcast: Play in new window | Download
32kbps Mono CBR (82 KB)
Podcast: Play in new window | Download
64kbps Stereo CBR (164 KB)
Podcast: Play in new window | Download
128kbps Stereo CBR (324 KB)
Podcast: Play in new window | Download
128kbps Mono CBR (324 KB)
Podcast: Play in new window | Download
256kbps Stereo CBR (647 KB)
Podcast: Play in new window | Download
Quality-5 Stereo VBR (295 KB)
Podcast: Play in new window | Download
My recommended settings for great MP3s from LAME and Audacity
Go into the encoder options and switch to “Variable (VBR)”, set the quality to “5: 110–170 kbps,” Variable Speed to “standard,” and Channel Mode to “Joint Stereo.”
Recommended LAME settings
Share your thoughts and experience!
What kind of settings have you used when encoding MP3s? How has your quality or file sizes changed since using my suggestions?
Need personalized podcasting help?
I no longer offer one-on-one consulting outside of Podcasters' Society, but request a consultant here and I'll connect you with someone I trust to help you launch or improve your podcast.
Ask your questions or share your feedback
- Comment on the shownotes
- Leave a voicemail at (903) 231-2221
- Email feedback@TheAudacitytoPodcast.com (audio files welcome)
Connect with me
- Subscribe to The Audacity to Podcast on Apple Podcasts or on Android.
- Join the Facebook Page and watch live podcasting Q&A on Mondays at 2pm (ET)
- Subscribe on YouTube for video reviews, Q&A, and more
- Follow @theDanielJLewis
Disclosure
This post may contain links to products or services with which I have an affiliate relationship and may receive compensation from your actions through such links. However, I don't let that corrupt my perspective and I don't recommend only affiliates.
Related Content
How did you make the mono examples? I only see Joint Stereo and Stereo as choices in your screen shot. Thanks for a great episode!
I do that before exporting. All of your tracks must be mono. You can do this by two different ways. Select all of your tracks, then go to Tracks menu > Mix and Render, but this will mix all of your tracks down to a single stereo track, which is destructive. Then go to Tracks menu > Stereo Track to Mono.
Or select your stereo tracks one at a time and click Tracks menu > Stereo Track to Mono. This is what I do on my vocal track, which doesn't need to be stereo and it cuts my uncompressed archive for in half for the vocal track.
Actually, I'll confess a mistake I had made when I processed this. In the episode, I demonstrate speaking in left and right channels. I did this by turning the Balance knob on my mixer while I spoke. But I had forgotten that I had done that, so when I processed, I still mixed my vocal track down to mono! I had to go back and artificially split the channels in that section. Hence the repost.
I wondered about how you did the left and right panning if your file was mono, as well as what the repost was for. I listened while I was out walking, and the info in the episode was compelling enough for me to come to the show notes to look it over again. Thanks for the great show!
I release my episodes in stereo, but I always mix my vocal track down to mono, just for space reasons when I archive to disc. I record in four channels with my H4n, so my vocals are always on their own stereo track.
@theramennoodle You discuss Mono vs. Stereo & your 'secret sauce' Audacity VBR settings. Are your shows VBR & Mono or not?
My shows are VBR, Quality-5, Stereo encoded from LAME. I use the secret same secret recipe that I shared in the episode. I chose stereo for the sake of my music. However, I am considering dropping back to mono with a lower quality (since Quality-5 is very high for just a single channel) to reduce download times for my subscribers, and make it easier if I every have to move my files to LibSyn.
I noticed that the example mp3 you played using your final "secret" recipe does not sound as good as the actual audio of your show that surrounds it. Is that because you re-compressed the mp3 again when you encoded the final mix of your show?
Yes, that is correct. I warned of this in my episode because I was recompressing already-compressed audio. This is why I advise everyone to distribute their promos in uncompressed WAV and get their sound effects in uncompressed WAV.
When i first started using audacity to record webcasts, a one hour webcast would take about 5 minutes to save as mp3. Now, to years later, it is taking 15-25 minutes. How can I get back to 5 minute process. ps i clear out audacity temp file daily. thanks
The higher your quality, the more time it will take to convert. What are your export settings?
What version of Audacity are you using now? What operating system? Are you on the same computer as before?
Audacity 1.26 on widows xp home . Yes, I am om the same computer doing the same things
how to change male song to female voice .I have tried but it crapped .
Hi! Changing a male voice to female is harder than going the other way. Listen to my episode about how to apply special effects in Audacity.
[…] podcast MP3sLAME is the best MP3 encoder! But it’s the best for music, and only when you use variable bitrate (VBR). This isn’t very compatible for podcasts, and using LAME any other way produces poor quality […]
[…] Normalization (I recommend a -1 value) on your sound clips to raise them to the same volume.ExportAudacity can make its own MP3s with LAME, but I don’t recommend it.LAME works best for music encoded at variable bitrate (VBR). This […]
This episode was extremely useful. It answered a lot of questions I have had for a long time. Thanks a lot!