My wife is really into audiobooks, and frequently has me rip copies of books on CD to digital files.
Years ago, so had an iPhone, so it made sense to use AAC encoded files in MP4 containers. Over time, I settled on using Foobar2k to do the encoding. After all of the discs were encoded to single files, I would use tagging software to add the album art and metadata.
However, this would still yield 5 to 15 individual files, depending on the book. Recently, I had obtained an audiobook copy of Ready Player One for my daughter from Audible. I thought it was convenient that the book only used two files, so I started looking at ways to streamline my process further.
Starting with the codec: Since my wife doesn't use an Apple product now, I was free to look at alternatives. It appears that Opus is the new sexiness when it comes to speech. Also, it is supported on every platform (except Apple... unless you put it into their Core audio container). A 24kbps Opus file sounds as good, or better than what I had been doing with AAC.
Next, I need to figure out a way to automate some of the disc juggling, and concatenation of files into two or three big files per book.
Here is what I tested so far: I used cdparanoia to rip each disc to a single file. After I get 5 or 6 discs worth of .WAV files, I used sox to combine those into a single FLAC file. I attempted to use wav at first, because it was faster. That is when I discovered WAV file size limitations. After the FLAC is finished encoding, I encode it into an Opus file, and finally finish it off by adding the metadata.
So now, I have six discs worth of content in a nice little 80MB file (or two files @ 160MB for the whole book). Not bad! If opusenc supported multiple input files, I could skip making the FLAC.