Converting Processed Sermon Files to MP3
In the previous article, I explained how to use Audacity to process a recorded sermon to ready it for podcasts and radio broadcast. The resulting WAV file is rather large, and cannot be conveniently stored or streamed, and so it should be converted to MP3 format. The LAME encoder is used to create the mp3 files for streaming and archival purposes, and we will use the AutoIt V3 scripting tool to simplify the encoding process.
For web streaming, the source file is run through a lowpass filter at 6kHz, converted from stereo to mono, downsampled from 44.1kHz to 11.025kHz, then the mp3 compression algorithm creates a 16kbps output file. That's a lot of technical jargon, but the point that this bandwidth is low enough to stream over a dial-up connection while minimizing the weird noises (compression artifacts) created by squashing the data into such a small file. Our streaming files are just over 1MB per ten minutes, as compared to 1.4MB/minute for archive MP3 and 10MB per minute for CD-quality WAV files.
Now, download this script, save it to the folder where you will be processing WAV files, and rename it from a .txt file to lametwice.au3 so it will launch properly using AutoIt. Right clicking this script in Windows Explorer should offer you the option to Edit Script, which will open the file in Windows Notepad. Do this, and change the $lameCmd variable to match the location where you installed lame.exe, and fill in the correct name for $defaultSpeaker.
Double-click lametwice.au3 and select your processed sermon WAV file. If you named the file correctly, then the first dialog box should already show the sermon title, though you may need to add punctuation.
The sermon date should also be filled in from the file name.
Your default speaker name will appear next, but can be changed for individual messages.
Finally, the script will prompt for a scripture reference. This is optional, and will be added to the date and stored in the MP3 comment field. Our podcasting scripts read this information and display it as part of the podcast file.
After all of the information is entered, a confirmation screen will be displayed. You can cancel from this screen to abort the MP3 file creation if anything went wrong.
The script will then launch the LAME encoder twice. The first pass will generate a low-bandwidth podcasting version of the message.
The second pass will generate a high quality file for archiving. This file is near CD quality, and should be sufficient for storage and for use in the radio broadcast, eliminating the need to store the full-sized WAV file.
After LAME completes, the MP3 files should be ready for upload and storage.