If you're an author, you know how important it is to have your book available in as many formats as possible. That way, readers can enjoy your work no matter how they prefer to consume it.
And with audiobooks becoming increasingly popular, it's essential to have your book available in that format as well. However, the thought of recording your own audiobook can be daunting.
Not only do you have to worry about the quality of your recording, but you also have to worry about narration, getting it right, editing, and so on.
We've all heard those unnatural, robotic voices that Ai audiobook narration can sometimes produce. But new advancements in technology are beginning to change that, and some Ai voice-generating software is now capable of producing strikingly realistic human speech.
Well, what if I told you that there was a way to clone your own voice and use it to narrate your book?
Interesting right?
But, how?
The Neural Text-To-Speech Method Nobody Is Using!
The technology exists, and it's called neural text-to-speech (NTTS).
With NTTS, you can take a short recording of your voice - just a few seconds - and use it to generate an unlimited amount of speech that sounds exactly like you. The longer audio sample you use, the more natural-sounding and more like you the final product will be.
Not only that, but you can also use NTTS to control the tone and pitch of the generated speech. So, if you want your book to sound happy and upbeat, you can make that happen.
Or, if you want it to sound more serious or dramatic, you can do that as well.
The possibilities are endless!
So, how does it work?
In a nutshell, NTTS uses artificial neural networks to analyze your speech and then mimic the way you speak. The software essentially learns to imitate your voice.
Don't get freaked out, it is not as intrusive as it sounds, and can help you clone your voice with incredible accuracy to the point where you could generate thousands of words of text just as if it were you recording it yourself!
That 80,000 word novel you've got there, saving up to afford an ACX narrator? You could be done with the narration in a few days, and all you need is a decent quality recording of your voice to get started.
NTTS is still in its early stages, but it's already being used by some big companies. For example, Google's Duplex system uses NTTS to make phone calls on your behalf and sound just like you.
I must admit, that is a little creepy...
But if you are still interested, I have been trialing this for a little while now and after testing out a few different services, I am happy with my final choice!
The Best Audiobook Narration Software For Authors
What is the best audiobook narration software for authors?
Descript has been marketed as a transcription software and has a tool for podcasters to correct their mistakes without having to re-record chunks of their episodes. The feature is called Overdub.
But, that same feature that allows you to train the AI to learn and clone your voice and pronunciation to correct audio mistakes, can be used to generate and edit massive amounts of text that sounds like you.
Most authors do not even know this is possible and have ruled out ever making their book into an audiobook as they think it will cost too much.
I have been able to generate thousands of words of text that sound exactly like me, and the best part is that it is all editable. So, if I make a mistake or want to change something, I can just go in and edit the text.
As mentioned, it is also possible to control the tone and pitch of the generated speech, so if you need to stress a particular point, you can.
Click To Watch: Overdub - Editorial Correction Demo
There are a few other options on the market, but at the moment, I believe Descript's Overdub feature is the best audiobook narration software for authors as it offers a lot of features and flexibility and doesn't break the bank in the process.
Continuous Improvements To Your Audio Quality
Another cool feature is the ability to improve your output simply by adding new audio clips of yourself! I prefer to record a dedicated audio clip of myself reading text as it allows me to pronounce things the way I want them to sound. The more data it has on your voice, the better the output will be.
This page offers chunks of texts which cover all the different sounds of our vocabulary. It is best to read from a document like this as it pushes you to say different words and sounds that may not be typically spoken in your day-to-day speech.
How Much Does It Cost?
It is FREE!
Yup, Overdub is free on all pricing plans, so you can try it out and see how your voice would sound. The only thing is, on the free plan, the vocabulary is limited to 1000 words.
This means you can still generate your text, however, you would need to upgrade to the paid plan to have those words outside of the 1000 word vocabulary limit. You will notice, on the free plan, words outside of this limit are replaced with the words "jibba" or "jabba" (which is actually quite hilarious).
The paid plans start at $12 per month, saving you 20% with an annual plan. So, it is quite affordable, especially when compared to hiring a professional narrator, which can cost hundreds or even thousands of dollars.
Is It Worth It?
I would say yes, absolutely! Because on top of narration, you have a plethora of other features you can use for all of your creative and marketing needs!
Realistic AI Stock Voices
One of the most impressive features of Descript is its use of Lyrebird AI to achieve the state of the art in voice synthesis. With Lyrebird AI, Descript is able to create realistic stock voices that sound incredibly natural. If you're not looking to clone your own voice, Descript has a number of very realistic stock voices to choose from.
You can also control the speed, intonation, and pitch of the voices, so they sound more natural.