True. My original intention, before discovering how common wonky mp3 audiobooks were, was to use the MKA container (I guess MKA is audio MKV?) to simply preserve the original audio data streams without reëncoding. However, since my script converts to WAV then reëncodes to 48kbps OPUS, I probably should have FFMPEG use M4B as the final output container for clarity.
Here’s my Bash script for combining a directory of
.mp3
files into a single.mkv
container usingffmpeg
. It also reëncodes the audio using the opus codec which can help smooth out weird encoding artifacts from sloppily created MP3 files. The script retains the original MP3 file partitions as seekable chapters named after the input file names. I find it useful for some audiobooks purchased from libro.fm which are only available in MP3 format.There’s already a container for audiobooks - m4b
Mka is excellent because you can add all the chapters and everything to the file and nothing needs to be re-encoded.
True. My original intention, before discovering how common wonky mp3 audiobooks were, was to use the MKA container (I guess MKA is audio MKV?) to simply preserve the original audio data streams without reëncoding. However, since my script converts to WAV then reëncodes to 48kbps OPUS, I probably should have FFMPEG use M4B as the final output container for clarity.