Photo by Eric Johnson, engineer at Clatter &Din studio

Image description: A woman sitting before a hideously expensive-looking microphone in a sound studio, wearing headphones, and reading from an iPad placed on a music stand which is covered with sound-deadening old carpet. To one side is a wheelchair and a small table.

In February I recorded So Lucky for Macmillan Audio. It was my first professional book narration.

Making it happen

From the minute I wrote the first draft of So Lucky I wanted to narrate the book. I like to read aloud. I used to front a band, and performing my own fiction has always been the next best thing to singing. Whenever a new novel is published I love to read from and talk about the book, followed by audience Q&A; it’s my favourite part of the publication process. I thought Lucky would be a great book to learn on: it’s mine, so I knew how it should sound and how to pronounce all the names; it’s short; and, given that the entire book was a complete surprise to everyone, no one had any set expectations. My editor has heard me perform before, and he was fine with it. It took a little longer to persuade Macmillan Audio but eventually, in February, everything was sorted and off I went to Clatter & Din, a beautifully wheelchair-accessible studio in Seattle’s SoDo neighbourhood, and laid it down in nine hours over three and a half days.

I wasn’t going into this completely ignorant. I’d done my homework and talked to professionals—Janis Ian, whose second GRAMMY was for spoken word narration of her memoir, Society’s Child, was particularly helpful. She talked to me about how often to take breaks, the best way to preserve and extend voice stamina, even what cough drops were most useful.1 I also talked to the producer at Macmillan, Matie Argiropoulos, and she said: Don’t worry, we have authors narrate their own books all the time.2 Just read over the first bit a couple of times aloud—though not at full volume!—and you’ll be fine. Oh, and rest your voice as much as possible.

That last part was what worried me. MS makes me a bit more liable to fatigue than most people, and performance—of any kind, but particularly voice—is notoriously tiring. Given the word count of the book I thought the finished recording would be somewhere between 4 and 4.5 hours (it ended up being 4:52, including credits). But how long would that take to lay down? How many hours would I have to be reading, reading, and reading again?

Cursory research gave me a consensus estimate of two hours per finished hour, or perhaps a little over. The more expert you are, the less time it takes. Matie, being sensibly cautious with a first-timer, booked five days in the studio.

Final pass proofs wouldn’t be done until the beginning of March, far too late for recording. So I made changes on a Word document, upped the font-size to 14-point for ease of reading, then turned that into a PDF, and uploaded it to m old iPad Air. I also sent it to Matie, who forwarded it to the sound engineer. Now we would all be reading from exactly the same document: literally on the same page.

The day before we were to begin I ran through the first forty pages out loud. (Mindful of both Janis and Matie’s advice to save my voice, I whispered rather than using full volume.) It was immediately clear to me that I’d have to add dialogue tags here and there on the fly. What’s obvious on a page is not obvious to a listener. I was used to adding tags in public readings but I wasn’t sure if that was a Done Thing for the official narration. I checked with Matie: Yep, that would be okay. And I could also add directions for things like texts, Tweets, and emails.

It begins…

On the day, Friday, I dressed in studio-friendly clothes (soft non-rustling fabrics, no metal zippers or dangly/jangly jewellery—none of which I wear in everyday life, anyway), loaded my wheelchair backpack with essentials (chamomile tea, water, iPad, two kinds of cough drop, notebook) and with Kelley plunged into the horrors of Seattle rush hour traffic in freezing temperatures. We arrived at the studio just before 9:30 and were greeted by a dog, Elvis, and half a dozen techs, engineers, and back office people with long hair, flannels, fleeces, hats and tats. We all chatted for a bit, explored the break room, got some tea. This was going to work.

The studio I was to work in was enormous, big enough for a large band and maybe a small choir. (Seriously: huge.) It was also cold, but that’s okay with me; I was dressed warmly and I’d be drinking mug after mug of hot tea. I transferred out of my wheelchair to a small, solid, non-creaky chair in front of a music stand covered in sound-deadening carpet. I put my iPad on the stand. Turned it on. Laid out my supplies on the small side table. Then I turned to the hideously expensive-looking microphone, pop-filter, headphones, and shock-proof stands, and then the straightforward-looking input board for the headphones. After a bit of experimentation I found the right heights and angles of chair, iPad, and light, put on the headphones, and started testing.

Eric was in the sound booth, visible behind glass, at the massive board. Kelley sat with her laptop at a desk behind him. (She’d planned to get some work done but in the end mostly listened to the narration.) They could both hear what I said via speaker; I heard Eric through the headphones. I also heard my own voice. And Matie, who was looped in via Skype from New York. It only took a minute for me to adjust the headphone mix to what was comfortable and for Eric to find a rich mix with plenty of range left for the bits of the book that get loud (and parts of it do).

Matie said Hey, asked if I had any questions, spoke about process for a minute—just read naturally; she would interject with corrections; she would also hear when I was getting tired and suggest a break. Then we chatted about accents. No, I said, I wasn’t going to try American accents, except for one old southern women because we need her to sound distinctive. (There are reasons. You’ll see/hear if you read or listen to the book.) She told me to just begin without copyright info or acknowledgements, so I plunged right in.

After about ten sentences Mattie stopped me. Relax. Breathe naturally; we have breath-scrubbing software to take out the worst bits. (I hadn’t known that!) Also, it’s a bit too fast; start again. Surprised—the audiobooks I’ve listened to always seemed to move faster than I would read aloud for an audience, so I’d assumed I should, too—I did. This time everyone was happy. We were off…

It took me about two pages to get the hang of just stopping when I made a mistake, backing up to the beginning of the phrase or sentence to match breath and tone, and doing it again. Every time that happened Eric made a note with his Pencil on his iPad Pro so that the editor, when they got the files, could see where to focus. It began to go very fast. So Lucky is a short book, and I’d been working on the proofs just the week before, so I knew its rhythms by heart.

Matie and I quickly found that we both preferred the same communication style: minimal, but swift and direct. No hedging. To some, perhaps, it might have felt abrupt or too blunt, but it worked perfectly for me: tell me once, clearly, then step back and let me do it. I liked it; I think our mutual understanding made life much easier. Of course, it could be that Matie has done this so many times she can adapt to anyone’s style, in which case, I owe her even more thanks.

We stopped for our first 15-minute break after just 40 minutes. Back again for 35 minutes. Break for 10. Back for another 30. Break for an hour’s lunch: fabulous Indian food and scritching of Elvis. Back again…and this time I started to cough. I tried everything—tea, water, honey, cough drops—but it was no good. We were done for the day. Matie said we were already ahead of schedule so no worries

Halfway home I stopped coughing. I regretted the fact that we hadn’t got more done but, eh, that’s just how things go sometimes.

Over the weekend I practised the southern accent for the old woman: it had to be creepy and distinctive, but not hammy.

On Monday morning I was fine. We arrived at 10 this time, to avoid the worst of the traffic, and started tearing through the text. Eric would stop me if some noise interrupted the recording (at one point I forgot to power down my wheelchair and the batteries in the hubs beeped as they turned off after half an hour; at another, there was some weird howl from the equipment that no one understood) or when he heard mouth noise. I’d sip some tea and go back a sentence or two. Mattie would stop me every now and again when either I made a mistake I didn’t catch—marble for maple, say—or she heard me getting tired. (I’d say that on average we stopped about every 35 minutes for about 10 minutes.) Very occasionally (I think it happened three times) she’d break in and ask for a different intonation. Things were going brilliantly, very fast and smooth.

eyes closed

The first day. I drank a lot of chamomile tea. Cups and cups and cups. Photo by Kelley Eskridge.

Image description: Woman in profile, wearing headphones and smiling as she sips from a mug of chamomile tea. Just visible on the left are the pop-filter of a microphone and the corner of an iPad on a music stand. Visible behind her are a wheelchair and small table.

We broke for lunch again. Back into the studio…and I started coughing again. And, again, I couldn’t stop. Again, Matie said: No problem, we’re seriously ahead of schedule; go home. Yet again, on the way home I was fine.

I thought about it that night and realised: it was Elvis. I’m officially allergic to dogs but I spend so little time around them and it’s such a minor allergy that I tend to discount it. But audio work is unforgiving.

The next day I had a plan: I would stay away from Elvis at lunch time, and take a Benadryl 20 mins before I got back to the microphone. That would give me a solid 40 mins or more before the Benadryl hit me like a sledgehammer and I start to get stupid and slow. It worked. This time we had a different engineer, and it was a bit less smooth—his style of recording, playback, and commentary were really different—but still went fast. The only problem was that my stomach gurgled a lot. No one was fazed by that; it happens at lot. Matie told me to hug a cushion tight against my belly which reduce the number of gurgles and lessened the volume.

Image description: Two photos of a woman facing the camera from behind a professional studio microphone set up and lit music stand, wearing headphones. On the left, she has that impatient, “Yes, can I help you?” look. On the right, she is smiling.

At one point, I’m sorry to say I don’t remember when, and I recorded the credits: Macmillan, Matie, me. It’s pretty weird announcing my own name as author, but at least I got to say “Narrated by the author” rather than having to say my name twice. Then we went back to the main text. And as I went I got faster and more accurate. If we’d pushed another 30 minutes I think we could have wrapped the book on Day 3. But Matie said, No, let’s come back fresh tomorrow and finish strong. After all, Macmillan had paid for the studio for five days. And maybe tomorrow when we were done we could get some photos for Macmillan to use in publicity? And also she wanted me to think about how I was going to play the ending.

She’s the director, and a professional, so I said I would. But I didn’t; I didn’t have to. I knew exactly how I was going to read it. This was the emotional crux of the whole story; there was only way it could go.

The next day we got there at 10:00 a.m. Yet another new engineer, and I sighed inwardly. This last bit of the book had to go exactly right and it would suck if this engineer was hard to work with. It did take us a moment to find our rhythm but then we did, and it was great. We began. And before I knew it it was upon me, the ending, and it poured out—it roared out—and it was perfect. It was done.

Everyone was silent—I thought maybe my headphones were broken—and then I realised that behind the glass the engineer was wiping his face and Kelley was grinning like a maniac.

How it felt

I have always loved to perform, and that momentary silence at the end from any audience means they liked it. But the exhilaration I felt this time went beyond my expectations. It’s hard to articulate the thrill as those final words tore from my mouth, clothed in the power of the human voice, and took form. They felt like living things. It felt like sorcery. I cannot wait for you to listen to this book, especially the end.


What will I do differently next time? The first thing is to buy an iPad Pro, a large one, and a Pencil, so I can mark the text with reminders about accents, emphasis, breath. Doing So Lucky without notes was doable, because it’s mine and because it’s short. But if I wanted to narrate, say, Menewood, the Hild sequel, the length and complexity would absolutely demand notes. And if it were someone else’s book, notes on the page would mean I’d have to spent less time practising.

Next time I’d also avoid dogs. And I’d take my own food to avoid allergies.

And next time I’d like to get the contract signed and sealed before the job begins (this one wasn’t signed until the end of February, two weeks after we wrapped). For So Lucky I had no leverage: I was untried, a beginner. There was no way to get more than Macmillan’s base rate ($250 per finished hour) and their standard boilerplate (though actually we managed to get some of that changed). After all, I wanted it more than they did. Also, they were taking a risk. I might be crap: they might have to eat the cost and start again with a real narrator. But now I’m a known quantity and quality, and it’s not crap.


Seriously, I can’t wait for you to hear this. It’s out May 15 from Macmillan Audio, but you can listen to an excerpt on SoundCloud. Then preorder here:

1 Janis recommended Pine Brother Honey Cough Drops. Sadly they didn’t work for me, I’m guessing because of the corn syrup, but the rest of her advice was spot on.
2 Turns out that Macmillan has nonfiction authors read their books all the time; fiction is less usual.