Full Version: Dramatic voice overs using AI and text to speech
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
I'm been messing around with some of the newer text to speech apps available and came across one that really stood out called Replica Studios.

It's different from the others because it focuses on "dramatic actor" voice-overs.

It doesn't have a lot of character voices at the time, I'm guessing around 50 and half aren't that good. However, the good voices are the best I've found.

Replica Studios is also fairly expensive when compared to other text to speech apps, but it has a "pay as you go" structure that's reasonable IMO. You get 4 hours of text to voice for $24. Let's do the math:

That's $6 per hour of voice over. For a 10-minute video it's $1 and for a 1-minute video it's 10 cents.

There's both a web interface and WIN and MAC desktop apps you can download. Firefox wouldn't play the audio previews for me on the web version, but was OK using Chrome.

The software gives access to a few more features like the ability to select text and change the pitch and speed of the voice. The software needs WIN 10.

Some of the voices have different "moods" like angry, happy, wise, etc., so check for those first when you need "main characters".

You can also use Replica Studio voices as secondary characters in your videos. For example you want two people to talk on the phone (or with each other). Or use a dramatic voice to create a "SUBSCRIBE NOW!" stock audio clip. Because it's text to speech you can have it say pretty much anything you want, with respect to the usage terms.

Which brings us to...the terms. They have a pretty liberal usage license including royalty free commercial use, but do read it as there are multiple copyright issues concerned. One is your text/content and the other is the real person behind the voice-over and Replica. And you can't use voices in negative/harmful ways. But there's a LOT a creative entrepreneur can do with them in honest, ethical and legal ways.

A quick example of a couple of voices:

Here's an example with background sound efx:

This is an example of what can be done using an available Replica Studio plugin for a program called Unreal Engine (free):
Replica Studio may not be the best text to speech for all situations, mostly because of price and not a big selection of voices.

But there are some really cool ways to use it and in many situations it only costs pennies to do amazing things. Having access to to "dramatic character actors" on demand is powerful for marketers. Once you learn how to do's mostly experimenting with the different voices to find the ones you like, you can create emotional voice-overs in minutes for a few pennies.

How about:

Create "batman like" sound effects like POW, BOOM, etc to use as "audio pattern interrupts" and call to actions.

Make voice-overs for video memes and quote posters.

Use as a narrator in short-content videos for a number of niches where the voice(s) are appropriate. Voice-overs for short videos costs 10 cents or less after your free trial minutes are used.

Spell out your domain name, a phone numbers, or create a catch-phrase or a "jingle" for branding. You can pay a penny or two once and use over and over.

Use with cartoon animation software like CrazyTalk to bring animated characters to life.
An interesting, and possibly creepy, feature is the ability to use your own voice to create a voice to use with text to speech. In theory, you could even use someone else's voice and fake what they are appear to be saying, and there are warnings about doing this.

While it seems cool at first, I checked it out a little bit. The first step to get going isn't too bad and you'll need to record your voice speaking some text then wait a few days for them to add it to their system. However, to get a really good end result you'll have to go through another couple of steps of reading and speaking text and uploading it to them. I "think" that last step is reading and speaking something like 150 pages of text.

An interesting benefit, if I understand this correctly, is that you can offer your voice-over for others to use and then get paid a percentage based on how much it's used. I wonder if the more popular voices are making any money?