Audiobook dictation apps

I’ve been considering recording my own audiobook, but having never done it before there are some things I’m not sure about.

First, the software. Currently I’m planning to use Audacity and just read it off from my Scrivener file, but I tried a bit of experimenting with that and Audacity turned out to be a little awkward. Are there any programs out there that assist with recording like that? Or do I just need to get used to Audacity?

Secondly, I’ve been told that I should get a condenser microphone, since the built-on mic sounds like garbage. I’m not sure if I should get a regular condenser along with a preamp to use with my computer’s line-in, or if just getting a USB condenser mic will work just as well.

Any suggestions?

I’d personally avoid using your computer’s line in for anything other than the occasional phone / Skype call.

Fundamentally, there are four main things to worry about when getting acoustic audio (such as your voice) onto a computer. In no particular order (they can all screw it up for you):

  1. Your environment: is your room neutral sounding - ie quiet (including computer noise) and free from too much echo / reverb (think furnished room vs tiled bathroom)
  2. Your ‘instrument’: is your voice nice to listen to? Breath properly. Practice talking into the mic to get the distance right. Make sure you are comfortable and able to maintain the same position relative to the mic the whole time. Get mic with a built in stand, or buy an external desk stand / boom stand. Consider standing whilst reading to help you use all your lungs while speaking.
  3. Quality of microphone
  4. Quality of A/D converter: this is what turns the microphone output (analogue) into a digital signal your computer can work with. Your line in jack on your computer is effectively doing this job but unless you have a pretty good dedicated soundcard (chances are you’d know about it if you did) it is likely to be pretty rubbish at this.

The two options available to you for figuring steps 3 and 4 are:
A) Separate A/D interface and microphone. This is what I use for my recordings. In my case it’s a Mackie Onyx Blackjack audio interface, plus a Shure SM58 microphone. But then, I’m a rock legend. This is likely to be overkill for your needs
B) Microphone with a built-in converter… ie one with a USB plug on the end of it. These days, USB mics are pretty damn good, and for recording voice more than up to the task. I’ve not tried them myself (as I use the above setup), but Blue Microphones do an well regarded range of USB mics at different price points that I’d certainly consider if I was in the market (check out, for example, their “Snowflake” desktop USB condenser mic).

For software, you are on a Mac, right?


I tried GarageBand, and I think it might work better. Thanks for the tips!

For the recording, I use Amadeus Pro. I’ve tried Audacity; it’s good but I find the interface much more confusing. Must say, I haven’t tried GarageBand. I opened it once, closed it immediately and went back to Amadeus Pro, professional, very good interface, just as powerful as Audacity but totally Mac-like and more intuitive.

And I totally agree with Pigfender on the rest, albeit from a Mac point of view.

For important recordings in China I used a pair of Sennheiser mikes — they’re over there and I’m in the UK but I think they’re e625S cardioid dynamic mikes, though Sennheiser seem to have replaced them with newer models. Which mike you use is up to you and your budget, of course. But go for cardioid — that means their sound capture is in a figure of 8 back and front in relation to the mike, cutting out extraneous noises from the side — or something directional.

The mikes were plugged into a small Behringer mixing desk, in my case UB802 — — to give me Firewire input to my Mac. There is a cheaper and simpler Behringer route for you, giving direct USB output: the XENYX Q502USB — one microphone jack — or the Q802USB — two microphone jacks. But there are many, many mixers/interfaces on the market.

Whatever you use to make the recording, the one thing you really need is a good pair of studio headphones. You need them to be able to hear all the glitches in the sound that you need to cut out or deal with. Again, mine are Behringer HPS3000 —

In terms of where you do your recording, PF is absolutely right … you need to find an environment which doesn’t have an echo — that was one of my problems, especially in the flat I moved to on the university campus, which was impossible. My current thoughts are that if I have to do more, I will cobble up a couple of open boxes lined with carpet — or perhaps with papier maché or exploded polystyrene egg trays or similar — into which each mike will go to eliminate echo from behind or the side, and create some sort of frame onto which I can hang a rug or some thick bath-sheets to eliminate echoes from behind me.

Another tip … for each mike, get a small round ebroidery frame, stretch over it a bit from a pair of old tights that you no longer wear, and find a way of hanging that an inch or two in front of the mike(s). You can then be close to the mike(s) while you’re reading, but it will cut out the popping sound that the sudden burst of air on word initial voiceless plosives — [p], [t] and [k] followed immediately by a vowel — produces. The commercial screens for that which I’ve found are horrendously expensive for what they are!

For you, a few of pieces of advice:

  1. Print out your text onto paper at a size you are comfortable with and in a serif font as they are easier to read — I’m happy with 12 point TNR, but you need to suit your eye-sight. It’s much, much easier to read from paper than from a computer screen.

  2. Print it so that each page ends, preferably at the end of a paragraph, at worst at the end of a sentence. Don’t have any sentences that run on over a page break. If you do it like that, as you read, each page comes to an end at an intonational break. You can then wait a second before moving your paper and wait a second to get settled, and then start on the next page. That makes it very easy to cut out the sound of the page turn as it shows up clearly in your editing software. If you have sentences breaking across physical sheets, it is much harder to maintain your intonation so you can cut out the page turn and have a seamless join as if you had read that sentence in one go.

  3. If you make a mistake, and we all do, stop, wait a moment, start again at least from the beginning of that sentence, or if it is practical from the beginning of the paragraph. Again, it makes seamless editing much, much easier.

  4. Do a trial recording of a page or two to get comfortable with the whole process. Most people are nervous the first time they do a recording, and that comes across as when we are nervous, our voices lose the lower frequencies. You need to get comfortable so you can read in the lower half of your vocal register, whatever that may be, because that comes over to the listener as relaxed and comfortable.

  5. While you’re recording, do it as if you are doing a reading in front of an audience, not to yourself and not as if you’re speaking to the microphones. If you read to your microphones or as if you’re reading out loud to yourself, it simply doesn’t come across. There are intonational/intensity differences between reading to an audience and reading to yourself or a microphone just in front of you; in the latter circumstances there is a significant tendency to drop the voice on the way — I’m doing some editing at the moment of recordings done by someone who does that all the time, and it’s the very devil … the recording almost loses the end of the sentences. So imagine there’s someone about 5 yards or more in front of you and read to that person. Keep your intonation up for descriptive and narrative passages; you can bring your imagined audience forward, as it were, during any dialogue, but how you’re going to deal with the voices of different characters is up to you.

  6. Don’t eat immediately before you do a recording, and only drink water. Eating, or anything with milk in it, coats your vocal folds and your voice will lack clarity afterwards.

Hope that helps.


EDIT: I forgot to say that, on reading speed, I found that using TNR 12 point (justified) on A4 paper with 2 cm right and left margins gave good results at 6 secs per line. If I used 1" (2.54 cm) per line gave 5 secs per line. I think if you read faster than that, again it sounds nervous and gabbling. :slight_smile:

Thanks Mark, that’s very helpful!

Amen to everything suggested by PF and Mark. A few other notes from years of theatre and professional broadcasting.

Try to forget the words are your own; follow the script. You don’t want, however slightly, to edit/revise, even in intonation, as you go along.

Unless you’ve done this before, learn to inhale quietly. That quick intake of air, one you hardly notice, can shatter a mood.

A particular second to Mark’s #5.

Decide ahead of time whether you’re going to present different characters in distinctive voices. The challenge increases by an order of magnitude for each new voice you introduce.

Listen to other audio books, comparing those read by the author with those read by someone else. A few writers read their own material so well – John le Carré comes first to mind – that they might have had careers as readers rather than writers. Many others do not.


I guess I’m going to rehash what other people are saying.

I had this idea myself a little while back. I got Blue Microphones Yeti, it’s a little on the pricey side, but it’s a great microphone. You might want to get a pop filter too with your Microphone if you speak close to it, I just recommend staying away from the Blue Microphone pop filter. The pop filter itself is great, but the attachment arm and clasp is a pain to work with.

As far as recording software, If you on Mac I’m going to have to say with go with Garageband. It’s more than enough tools to handle voice recording and podcast like production.

If you like me, your going to have to fight the urge not to whisper (afraid other people will overhear, yet I’m recording this for people to hear, but I don’t want them to hear me making it) or go on in a monotone voice.

You might want to look around the internets for some voice acting tips. Some of them aren’t so good, but try them out with the Microphone on, then listen to yourself. If you like it, use, if not, move one to the next one. I know most people can’t stand to listen to their own voice recording, but if you can’t bear to listen to yourself, I doubt anyone else with either.

And of course, have fun.

Thanks for the suggestions. I’ve been reading my books aloud to myself for years because it helps me catch things that I otherwise wouldn’t, so I thought, why not record it into an audiobook if I’m doing it anyway? But yes, I will definitely need to practice to get the voice right for an audiobook. I haven’t entirely decided whether or not to go ahead and do it, but I’m definitely leaning toward it. Thanks for the help.

Just another thought, Amadeus Lite, at USD 24.99 might be enough for you. It lacks some of the bells and whistles — seems to me adding more tracks and some high-end professional features might not be necessary for you. Could be worth looking into.


I produced people reading scripts and books for radio for decades.

I believe you need two things: a recording studio and a producer.

The studio will have all the gadgets you will require, and with any luck someone who knows how to use them. You may find that your local community college or amateur radio station has a spare studio for an hour or two per week.

Then you need a producer: someone who has been professionally trained to turn an amateur reader into a professional one.

Good luck.

Yup, I should point out that the Garageband recommendation was based on two things:

  1. It’s free
  2. It’s already installed on your machine.

A lot of people find it to be sufficient for their needs, but it’s not a “pro standard” DAW. That said, if your needs are straight forward (Get it into the machine, basic cut and past editing, basic eq adjustment) it’s probably up to the task.

But what would I know? I’m a Windows guy so have never used it.

Oh and Mark… D’oh! I forgot to mention the headphones. Yup. Really really important. Just as important, quality-wise, as the microphone.

Another tip…

Have you ever noticed that after a short period of time you no longer notice the smell of your own aftershave / perfume? Your ears do exactly the same thing.

So when you are listening back to your stuff during editing, take regular breaks to “recalibrate” your ears. You’ll be much better at noticing whether you start to sound tired in certain sections, or if you’re subtly changing tone / timbre as time goes on.

I think I’ve decided to go with Audacity. It’s not the cleanest interface, but it seems to work better for me than GarageBand. Of course that might change once I start recording for real…if I do. Time, and school, will tell.

My personal experience with Audacity on Windows is that it works reasonably well. Not as slick as paid for programs like Sonar and such the like, but there are plenty of tools in there.

As long as you can get the latency compensation set up properly, you should be fine.

As I said, I have nothing against Audacity other than its to me overly complex, somewhat confusing and non-Mac-like UI, but then it is open source and the UI will probably be more familiar and intuitive to Linux and perhaps Windows users.

In terms of its features and capabilities, it is obviously very good. And it is free.