Okay, I've done audiobooks in MP3 in the past, but would like to take advantage of AACs bookmarking with the iPod now.

Do you have any recommendations of encoding spoken work/audio book recordings? 32, 48, 64 kbps? Have used LAMEs "--alt-preset medium -a --lowpass 10" previously.

Thanks in advance.

not that it means anything definative, but just as a reference, itunes encodes their audio books (or at least the ones i have) at 32kbps, 24,000kHz, and in mono.

Encode with Nero HE-AAC with the "tape" setting

if its bad, then go for "radio"

"streaming" is around 64kbps and the result should be very good for voice

EDIT: do ipod support HE-AAC?
Is HE-AAC backward compatible......? From what i've read normal aac player can also play he-aac but at half of the sample rate..is it like that?

That bookmarking system is pretty neat, but what few know is that it even works for regular music. Just rename any m4a to m4b, and you can jump right into where you left off. It's great for long classical works and it syncs both ways with the iPod.

Thanks for the posts. I should mention I'm on OS X, so until Nero supports that, unfortunately it's not a possibility. Thanks for the suggestion though kotrtim.

Also about half the audiobooks I'll be working with are to transcode. Want to get it out of cripple stage as soon as possible. So VCSkier you're on the right track there.

I was thinking 48kbps mono or 64 kbps stereo. What do you think? It would give a little headroom for transcoding (hopefully minimize artifacts), but also room for music (which occasionally shows up). Both rates seem a bit high though as they're pretty much what I was using on MP3. Just hoping to tap any experience around here.

Hello all!

For books in MP3 I've used "--preset fast voice" setting, but now that I have iPod could you please advise me of similar setting for AAC, and which encoder for Windows would give the best quality for this job.

Thank you!

I don't have an iPod, however I think I have read that only LC-AAC can be decoded properly on the device. SBR and/or PS can't be decoded - yet. That may significantly limit how low you can encode your voice material properly.

For playback on a PC, I have been looking at the latest Coding Technologies aacPlus v2 encoder for low bitrate voice.

I have done some basic "voice only" tests to see how well it can perform at extremely low bitrates. I have not experimented with transcoding the material, but instead I used full frequency voice samples encoded from wav files.

Acceptable speech results can be generated as low as 18kbps (mono). This may be a little low if you are transcoding your material, but experimentation should yield you the best choice for your needs. Instructions and downloads to encode with CT encoder with Foobar are here on the forums.

That probably doesn't help much, other than to suggest that the CT encoder may be able to generate acceptable LC-AAC for voice. Have fun trying it out.

Yes, an iPod can play HE-AAC files but not to their full potential, kinda like mp3 players that don't support mp3pro.

As for transcoding (or ripping) audio books, I would go with the 64kbps bitrate in iTunes. From my experience ripping a audio book CD, I found that 32kbps mpeg-4 AAC was pretty good but my audio book also contained live speeches. The background noise of the crowd was just terrible, so I ripped at 64kbps and that pretty much solved the problem.

I would try 32kbps and if it is not enough then go up to 64kbps.

I encode my Audiobooks using iTunes with the following settings:

All Spoken Word
-- Import Using: AAC Encoder
-- Settings: Custom
-- Stereo Bit Rate: 64 kbps
-- Sample Rate: Auto
-- Channels: Mono
-- Use Variable Bit Rate Encoding (VBR): Unchecked
-- Optimize for voice: Checked

Mix of Spoken Word and Music
-- Import Using: AAC Encoder
-- Settings: Custom
-- Stereo Bit Rate: 96 kbps
-- Sample Rate: Auto
-- Channels: Mono
-- Use Variable Bit Rate Encoding (VBR): Checked
-- Optimize for voice: Unchecked

By the by... if there is someone with a more technical knowledge of what happens when you check the "optimized for voice" button, I'd be interested to know what the encoder is doing. Right now I'm blindly using it becasue the resulting files sound better to my ears.

Be more aggressive in detecting speech harmonics and allocating bits to them.

Since this thread of mine was resurrected recently, thought I'd mention there's support for bookmarking now with MP3 in iTunes/iPod also.

Yeah that's true, but neither the iPod nor iTunes will recognise the file as audiobook unless it has the .m4b extension.
I even tried using YAMB (MP4Box) to wrap mp3 into mp4 and then changed the file extension to .m4b, but iTunes wouldn't even add it to the library.*

If anyone knows how this might be possible that would rock! Clearly iPod will play mp3, and will also bookmark it, it just won't file it as an audiobook.
Is there a way to fool it?

I only ask because AAC encoding is sooo incredibly slooooow! It takes ?x longer than any other lossy format I know of.
If we could just reprocess the headers, or wrap the mp3 in mp4 clothing, whatever... much quicker!

Excuse me if I'm talking complete gibberish, I have very little understanding of how these things work. I just know what I wish they could do.


*Why doesn't iTunes give you some sort of notification in this event e.g. 'Import failed because...' or even '437 audio files imported successfully', rather than making you search hi & lo to see whether it worked or not?!! It's the only software I know that doesn't tell you WTF is going on! Drives me bloody crazy!

Sorry, pet rant over ;-)

I really don't understand how can strong compression artifacts be considered "acceptable" anywhere, including spoken word. Today when storage is so cheap, how can anybody live with quality of a tape recorder.

In the days I used to rip theatrical performances from radio, I recall LAME produced about 90 kBit/s of VBR in mono. I consider the material worth a normal bitrate.

Forgive me if this post seems too emotional.