It was meeting with readers of Joe Faber and the Optimists that first got me thinking seriously about producing an audio book. Many stroke survivors find physical books difficult to handle; my husband’s one of them, and only reads on a tablet now. For others, though, brain injury has affected the ability to read at all, and audio is their only option. So, alongside the new book I’ve been writing, it’s become a passion project to narrate and record Joe Faber.
There’s a lot of noise around audiobooks. Audio is big and growing, and indie authors are curious about jumping on board, because it looks so possible. Platforms like ACX and Findaway Voices can help you find a narrator; AI may, in the future, offer a way to get narration done on the cheap; while digital technology means that it’s never been easier to record yourself, at home. However, remuneration is poor and opaque, with Audiblegate rumbling in the background. The game’s probably not worth the candle unless you’re already selling big numbers, have a publisher to take the whole thing out of your hands, and have an adequate marketing strategy.
You can do all sorts of things, but just because you can doesn’t mean you should. If something is easy to do, it’s also going to be easy to do it badly.
Alongside the altruistic motive for creating an audio version of Joe Faber and the Optimists, I have an equally specific, but ego-driven reason for wanting to do the narration myself. Modest as I am, I can’t believe anyone else can read my work better than I can. I know where the jokes are, for heaven’s sake! I know the serious from the tongue-in-cheek; I hear where the emphasis of each sentence falls; and if I’ve used a word, I damn well know how to pronounce it.
It gets worse. I also have a very specific reason for wanting to edit the audio file myself: I don’t know anyone I’d entrust it to. (This person must exist, but we haven’t met.) I know how long I want the pauses to be. I know what needs space round it, what needs to speed up. Comic writing implies comic performance, and comic performance relies on timing. And voice. The readers who ‘get’ what I’m doing tend to say they like the way I write. There’s a definite Alan Bennett / David Sedaris factor at play, a whimsical, performative side, and any author who writes like this will want to capitalise on whatever’s idiosyncratic about their voice in the widest sense. Not something to be handed over to a robot.
Right. We’ve established that I’m a delusional narcissist. Read the diary below if you’re a fellow-author curious to know what happened when I tried. If you just want to cut to the chase – because seriously, some people are only reading this post to find out what happened in the end – listen to this. And prepare to be confused. Suffice to say, I should have started where I finished up.
Diary… Five acts, but not a tragedy
Stage one Decision made: I can’t justify the expense of hiring a studio and sound engineer, when I don’t expect to make any money out of this. Nor do I have the voice training to talk non-stop for that amount of time, days on end. I sign up for an ‘Audiobooks made easy for authors’ course for a modest fee, and order the recommended kit.
Two of the recommendations on this course proved excellent for my situation. Firstly, use a dynamic microphone, not a condenser mic. (ACX and many ‘how-to-ers’ are axiomatic about the need for condenser mic, but a condenser mic will pick up the sound of dust settling. With a home studio, getting rid of unwanted noise is the most likely thing to defeat you come the editing / mastering stage.) Second, the free software, Audacity, is all you need. (It’s widely used by language teachers. And why pay to be confused?)
The course’s technical recommendations for editing and mastering, though, turned out to be inappropriate for narrative fiction, and for my voice. Worse, some would inevitably lead to your files being rejected (e.g. pasting in silence – instead of room noise). Of course, you know nothing of this when you start.
Stage 2 I’m getting into it, I’ve organised my home studio adequately, but there’s one huge practical problem. The technique I’m being taught is to record a small amount, then go back and edit immediately. But my brain can’t flip-flop between performance and editing. Again, it might work if I were recording the shipping forecast, but not a story.
Direct speech proves particularly tricky; I need a micro-pause to adjust my voice, which then needs to be edited out, if the whole thing is to flow naturally. And I start to notice mouth-clicks… aargh. Once you hear this, you become obsessive. On the performance side, I get some useful advice and feedback from a voice-trained friend, including warm-ups. Rappers on YouTube are good on this, too.
I google everything, find good and bad on YouTube, and discover the de-clicker plug-in for Audacity. Plug-ins! Yey! I notice, on the dashboard, Audacity’s ‘ACX check’ – which checks overall sound quality for upload to Amazon – and think I might be getting somewhere.
But every recording fails to pass. On the technical side, I’m getting a sense of what’s wrong but don’t know how to put it right, so I talk to a friendly physicist. The method I’m using is doomed.
Stage 3 I’ve plugged away and recorded many hours, but if editing is fraught, mastering to pass the ACX check is going to be worse… AT LAST I do what I ought to have done in the first place and go back to the Audacity manual, which, lo and behold, has a whole section on mastering for audiobook. Nothing like the course I paid for, b.t.w.
As I edit, I’m confused as to the order in which things need to be done, so I post my query on the Audacity forum. They have a simple mechanism for uploading a 10 second sample, which I do. At this point the tide turns…
I get expert help from someone on the other side of the ocean who knows what he’s talking about. Koz is a legend in my life. He patiently takes me back through basics. We message to and fro on the forum over many weeks. I glean that he’s an experienced sound engineer and full of wisdom. ( I’ve always loved working with people who have any sort of technical expertise – especially expertise I lack – so it’s a nerdy, happy place for me.) (OK, a rabbit hole – a more sensible author might have said, stuff this, it’s all taking too long.) Koz went the extra mile. His last communication was along the lines, you’ll go back to chapter one and do it all much more naturally now.
None of this good stuff cost me a penny. Audacity epitomises the best of the digital age.
Stage 4 Big realisation. I fundamentally dislike the way I’ve dealt with my main protagonist’s direct speech. His speech is affected by the stroke – but I’d made him sound irredeemably grumpy instead of the bundle of wit he is. The answer is to hold one hand over the side of my mouth, mimicking the effect of the stroke, while allowing a smile back into the voice. That’s going to be a lot of re-recording… I decided to park this project for a while and focus more energy on the other work-in-progress. Writing. And yes, start all over again at some future date.
Stage 5 With the second draft of the new novel sent off for editorial feedback, I sign up with Findaway Voices. My aim now is to produce just one short story as a sort of experiment or ‘apprentice piece.’ I’ll go through all the stages and establish the best workflow for me. The project indulges my inner nerd. I keep a detailed spreadsheet, with every clip, every action, and key metrics logged. I work out by trial and error what to do and what order to do it in. The uploaded file passes quality control first time, and some weeks later it’s available for distribution. It’s certainly no worse than many audiobooks (hear a sample on Spotify – here).
In hindsight, what the dodgy course did achieve was to sucker me in to trying. I needed a bit of hand-holding to get started, and some basic terms of reference. But as for the promise that, once you’ve done this course and mastered the techniques, you should only need five minutes work for each minute of recording time… Yeah, right. If that’s your aim, leave it to AI.
All I need now is a few weeks with goldilocks weather conditions (warm enough for no heating-related noises, cool enough to keep the window closed) and nothing else to be obsessing about. Big ask…