Over compression kills podcasting

One of the great benefits of podcasts is that they enable you to listen to presentations from shows like SXSW when it's physically impossible to attend. Something I find really frustrating is when an audio file is so heavily compressed it causes the speaker to end up sounding like they were presenting underwater.

The problem becomes worse if a presenter isn't speaking correctly into their microphone resulting in the original audio signal having level fluctuations. If the audio file isn't run through a compression utility before it's converted to MP3 the compressed output can be so distorted when the signal drops that the words can't even be understood.

Having worked in eLearning for several years I've released a lot of Flash based content with streamed voice over. The problem we had with audio was that it had to be compressed enough to allow the visual presentation to work flawlessly, but at the same time sound good enough to not cause aural fatigue in the user.

After a lot of experimentation and user testing we opted for compressing the audio to 22,050 kHz at 16kbps. This setting offers a high enough quality for learning material and a small enough file size for streaming. For Podcasting purposes I wouldn't advise compressing your audio files to a lower quality than this.

Luckily if you're at all unsure about compression, you can simply use iTunes to prepare audio files using the 'spoken word' podcast compression preset.


Tags: Podcast Compression iTunes brand