My Current Audiobook Process

I’m putting this here mostly for me because after recording four non-fiction titles and six short stories I’m about to go dive in on a writing project. (Also, if anyone “listens” to books on YouTube can you drop a comment and let me know. I’m thinking I’ll put the short stories up there at least, but just not sure how much to prioritize doing so. Right now it’s a backburner project because even though I expect people to just listen even though they’re videos I want the closed captions accurate and that takes a surprising amount of time.)

Anyway. In case this helps anyone else, this is not a perfect process but it’s what I’ve come up with to get a product I’m happy with but not bog down too much too early.

  1. Before each session, record a “testing, testing, testing” bit to make sure that my audio is set to the right input device, that I’ve turned on the audio interface, and that I’m hitting between -10 and -20 db. (All issues I had at one time or another.)
  2. Record audio using Audacity with no headphones on. Save file using Raw, chapter number (if applicable), and project name. (e.g., Raw 1 Introduction)
  3. In Audacity, edit file to remove long gaps, repeats, and re-dos. Do not wear headphones. Save file as First Pass, chapter number, and then file name. (e.g., First Pass 1 Introduction)
  4. Export as .wav file.
  5. Import .wav file into Reaper. Apply pre-set FX Chain which includes Waves NS1 Mono, ReaEQ, ReaComp, iZotope De-Click, JS: De-Esser, and iZotope De-Clip.
  6. Listen to audio and adjust the threshold setting for ReaComp until I’m compressing somewhere between -4 and -6 for most of the audio.
  7. Save as Reaper, chapter number, and then file name. (e.g., Reaper 1 Introduction)
  8. Render mono version of .wav file. (If there’s a way to make this the default in settings, I haven’t figured it out yet so I always have to change from stereo to mono.)
  9. Open .wav file in Audacity, select all or 20 minutes for longer clips, and run ACX Check (under Analyze). Look at RMS level and figure out amount to adjust to get to -23 and then add .5 to that amount.
  10. Go to Tools, Macros and change the value for the Limiter to that amount in the macro I have that applies a Limiter and Normalizes the clip to -3.1 peak level.
  11. Apply that macro to the whole clip.
  12. Run ACX Check again to make sure it worked.
  13. Save file as Final, chapter number, file name. (e.g., Final 1 Introduction)
  14. Export as MP3 file. Make sure file name is what I want the chapter called when loaded to Authors Republic. (e.g., 1 Introduction) (ACX will import chapter names from your ebook file, but Authors Republic uses the name of the file you provide them. Be sure to keep numbering in there so that files are listed in order on Authors Republic.)
  15. Listen to file with really good headphones on and read along in book to confirm text. Note timestamp for any issues that need fixed like sounds that need to be removed, duplicate text that wasn’t caught, or words that were wrong.
  16. Re-record if needed. Go to First Pass version in Audacity if text needs removed. If multiple cuts need to be made work from the end of the file backwards.
  17. Otherwise go to Reaper version and click on trim envelope and select Volume (Pre-FX). Find each spot where a noise was noted that needed removed and use ctrl + mouse to manually draw it down until no longer audible.
  18. Export from Reaper as .wav and reprocess with Limiter and Normalization in Audacity. Re-export as .mp3 file and listen to make sure all changes worked.
  19. For the file edits above, save over old versions as needed. With the final Audacity file, that will need to be deleted first and then a new file with the same name saved.

This is probably not the most efficient process. I could likely figure out how to do everything in Reaper, but I’m much more comfortable cutting sections in Audacity. Reaper and I have time selection with a mouse issues.

Also, I like the ACX Check in Audacity even though it is not perfect and told me I was fine a few times when I was in fact off by .3 or less.

It was taking me a lot longer to process files early on because I was getting caught up with mouth or background noises during my first pass edits. That’s why I don’t wear the headphones at that stage, because I get distracted and want to start fixing things that the software will fix for me if I just let it go.

I also at one point was manually fixing clipping in my audio track by bringing the pre-processed sound down at those points but finally add the de-clipping tool in there to do that for me. It was fine when I clipped once or twice, but I then had one with 55 clips in it and that was not going to be fun to fix.

(Also, as I get better at not hitting first words in chapters or sections too hard that becomes less of an issue.)

I also think I could have for new recordings adjusted my settings for the microphone input to get rid of any clipping, but I think that would’ve also meant a higher adjustment when I used the Limiter. (I think, don’t quote me on that.) So I’ve struck a balance there.

As you can see above, I did decide to go with paid software and paid tools after trying the free route first. I was able to get the de-clip and de-click software a few versions back from the current one off their website at a decent discount so it seemed worth it. All told I think it was about $150 for all the software I’m using.

Also, doing a good recording is essential. It saves so much time if you get that right up front.

I’ve been lucky to catch a lot of issues as I record, so I just re-do a line right then rather than have to catch it at the end of the whole process. I am not, however, comfortable enough to not listen through during that first pass stage. I do know of experienced folks who will just make some sort of loud noise at any point where they re-do a line so they don’t have to listen through the entire thing in the first pass stage. I’m just not there yet. But I can see maybe getting there when I try to do the cozies which are 9x as long as each of the short stories I’ve done so far.

As for prep, my current process involves taking a decongestant before I start, chewing some gum, and then having water with apple cider vinegar that I drink while recording. (My dog freaks out and needs to go sit outside in the grass for half an hour before she’ll let me record, so it works out.)

As I record I also really try to pay attention to any gumminess in my mouth or any spit bubbles (gross, I know) I notice so I can just re-do that line immediately.

I also have to be careful that there isn’t something really loud that suddenly starts up in the background. I have a pretty good little space set up right now and a very forgiving microphone that doesn’t pick up everything around me, but I had to redo about five minutes of one recording that sounded like static in the background. I think because someone was mowing right outside and I didn’t catch it. I wasted a ton of time trying to get rid of that with processing when it ultimately was easier to just go re-record the clip.

Alright. Off to do some writing before I circle back to audio again in a week or so.

Another Audio Update

First, for the non-fiction readers out there, Regulatory Compliance Fundamentals and How To Gather and Use Data For Business Analysis are now live in audio. The links above are to the Books2Read page where I added four audio storefronts for each one, but if you have a favorite place where you like to listen to audio it should make it’s way there eventually or may already be there.

Surprisingly, I actually have some listens already according to the ACX dashboard. We’ll see if they last since ACX takes back sales sometimes. (If you listen to a title and liked it and are one of those people who return your audiobooks, please wait a week on Audible before doing so.)

But, regardless, that was a nice surprise that people had found the books and were listening to them already without any promotion by me.

Especially the data one that I’ve always thought had good information in it but just never found its audience.

I also recorded Data Analysis for Self-Publishers and Sell That Book on the non-fiction side but haven’t submitted those yet. Probably today.

This week I started to work on some fiction titles and that was interesting. Because I had a bit of an epiphany.

A long time ago I took a craft course and the instructor said that the power of writing is that it lets you be inside the character’s head. TV and movies are external. You hear the character and you see them, but you don’t get that internal thought process. (Usually. Some shows are set up to provide that, but mostly we watch from the outside when we watch movies or TV shows.)

But with writing, you can see what someone really thinks or feels, not just what they show the world. (Now, I could argue that what you’re actually getting is the narrator’s or character’s spin on things, but let’s not go there.)

The problem with writing is that the reader has to be the one that layers on the emotions and interprets the words.

Each reader brings their own world view to the page. So something I as a writer might think is sad or angry or tragic or ironic may not land that way with the reader if their life experience doesn’t mirror mine.

This week I’ve been doing short stories I wrote almost a decade ago. The first one I tackled was called The Price We Pay.

This story was very personal to me. It was born out of my dad’s struggles with having a terminal illness and how uncaring the world is about people like him. And about struggling financially. And how hard it is to be there for your family when you’re barely making it through each day. And about having a complicated marriage with someone you desperately love but who might leave you given the chance.

I have friends who wouldn’t connect to any of that. They’ve never been seriously ill or known anyone who was. They’ve never struggled financially. They’ve never loved someone in that complicated, messy, painful way.

So for that person reading that story, there’s no emotion to the words. There’s no resonance to the experience of the character for them to draw on.

(Interestingly, I think now with COVID it’s probably a story far more readers can connect to than before.)

The other story I did this week was Death Answered My Call about a woman who is essentially using prayer to keep her husband alive even as it becomes more and more clear that she needs to let go. (In the story it’s a magical red leather bag and a repeated mantra, but basically it’s prayer.)

I still remember with that story having a beta reader who didn’t understand the story at all. That beta reader was religious and fully believed that we go on to somewhere better than this so didn’t understand not wanting to lose someone and trying to keep them here.

Whereas for me, I wrote that story because even though I am not a religious person I prayed for my dad every single time he was hospitalized when I was growing up. He was religious and I figured it wouldn’t hurt to ask his god to help him out. I prayed for him right up until that last time…

(I also had a friend who thought the character got what he deserved because he was a skydiver so had no sympathy for either character. Sometimes having people beta for you reveals really interesting things.)

Getting back to the point.

For me, reading that story, all the emotion is there. I know what it’s like to want to keep someone and to finally have to let them go because they’re suffering too much.

Which is what made narrating those stories such a great experience. Because I was in control of the emotion behind the words, not the reader. I could deliver not just the words, but the emotions that go with them.

The Price We Pay was perfect for audio for that reason. It has a lot of internal thoughts and feelings–the story is probably half just thoughts– so it wouldn’t make a good television episode, for example, but being able to narrate it also let me give the story all the emotion that you can get from acting that maybe readers wouldn’t be able to bring with them.

I don’t think I realized until recording that story for the second time (I cried during the first recording–I hadn’t read that story in probably five years) that this unique combination of written word and performance makes audiobooks their own unique performance medium.

I had always up until that moment viewed audiobooks as just a spoken version of a story that was convenient for people to listen to while driving around or doing the dishes or whatever.

And who knows, maybe that’s what people actually want in their audiobooks and they’ll hate these recordings. Maybe they just want that straight delivery and not actual acting.

Plus, I could suck as an actor. I mean, really, my acting experience is a middle school radio skit, a drama class I dropped in high school, and a freshman play in college, so it’s possible they’re horrible and overacted.

(If so, I’ll figure that out about two years from now when my mind finally finishes processing whatever it’s always processing in the background and makes me circle back to these and cringe. That’s life as I know it.)

But I also think my own personal writing style lends itself to audio.

I tend to write short, choppy sentences and paragraphs. Because everything I write I’m reading out loud in my head. And line breaks for me are a longer pause than a period. I see a long paragraph and I think, “Didn’t you take a breath in there, man? Geez.”

So maybe audio is just my natural medium but wouldn’t work for others?

I will say though that it is a level of difficulty or ten beyond just writing a story.

There’s the recording environment. The proper equipment. The sound processing. The actual vocal performance which can be impacted by energy levels and what you ate for breakfast and whether it’s allergy season and whether your stupid stompy neighbor is home and whether your dog is going to be okay with you disappearing behind a moving blanket to make weird noises.

And that all on top of having good words to work from in the first place.

A ton of things need to come together for a good audiobook recording.

I have no idea if I’ve brought those together. I think what I’ve done sounds good. But, you don’t know what you don’t know.

(As an aside. The audio I recorded for my course videos in the past, the issue there was that I didn’t notice there was a “distance” to the vocals. It’s like I’m across the room, not right there. They’re still workable for delivering that knowledge, they’re just not “polished.” Then again, that’s kind of the story of my life. So, what’s new?)

Alright. Anyway. Time to wrap this up and do some covers and then submit some audio files. I’d like to do all of the short stories in my M.H. Lee short story collection, but I don’t know if I can actually pull off narrating all of them. It’s going to be interesting to try.

Some More Thoughts on Audio

I just finished spending about an hour recording audio for one of my books. I have two I’ve finished and submitted for approval and two more non-fiction I want to do before tackling any fiction.

One of things I was thinking about today is that audio and writing are very different in one key aspect.

With writing you can be having a bad day, feel tired, feel out of sorts, have a head cold, what-have-you. And you can still sit down in front of your computer and put new words down. Maybe they don’t flow as well as normal. Maybe they need heavier revision when you come back to them.

But you make progress. Even if all the words are junk and you went in the wrong direction, it will often help you figure out where you do want to go with the story.

So on the writing side, pushing through is what you do. Butt in chair, right? We don’t always come to the keyboard full of joy. Some days we just know we have to keep going.

Audio is the exact opposite when it comes to this. Because if you record while tired or with a bad head cold or while too distracted to deliver any emotion to what you’re reading, it’s there in your voice. And you can’t edit that.

A flat performance or one full of stomach noises can’t be fixed. It needs to be re-recorded.

At least if you want a solid performance.

So with audio it is in fact possible to waste time on bad work.

I suspect this is one of those things where the more you do it, the less bad takes you record, but just starting out know that’s an issue you’ll be facing.

And if you have a bad take? Admit it early, throw it out, and redo it. Because five minutes of audio recorded poorly costs ten minutes of time at most. But if you edit it and process it and all that? That’s more like thirty minutes spent on something you’ll still have to redo in the end.