Voice over online free can sometimes be extended to certain nonprofit groups or for close friends and relatives of the artist. It is generally something that is not widely available to the public. I’m not opposed to assisting charities by providing voice over samples.
You could register as a client or buyer on sites like Fiverr and post that you are seeking a low-cost or no-cost voiceover. While you may not get a lot of responses, new sellers on the platform may be willing to help you out in order to establish their rating history on the site so they can attract more and paying clients. It’s often a win-win situation.
Some people use Google Translate for text-to-speech (TTS) voice over samples by pasting the desired text into the box and then recording the spoken translation. You might check into this further, or have your attorney look into it for you to make sure there are no legal ramifications for using their services in this way. I did see a portion of Google’s Terms of Service which reads, “You may not copy, modify, distribute, sell, or lease any part of our services or software.”
While this article makes it appear that this practice is permissible, the post is dated 2016 and the website’s disclaimer says the information is for “educational purposes only” so make sure everything is on the up and up if you are considering going this route.
There are a few TTS Synthesis platforms which offer voice over online free. MARY requires a download and as you can hear by this demo it sounds a bit garbled. A clip of Edgar Allan Poe’s “The Raven” poem on eSpeak’s website sounds even worse than that. It is completely robotic sounding with no pacing and the voice is very flat.
I have heard one artificial intelligence (AI) TTS offering that initially sounds decent, but the speech was not paced very well, the phrasing was off in a few places and the voice trailed off to where you couldn’t hear it clearly. Also, background music was used, most likely to cover up other flaws that weren’t so noticeable.
While some believe AI will replace human speech in this field, experts state that it would take countless hours of programming to create the depth, inflection and emotion of a real human voice and that would not be cost-effective. Plus, one thing they will never be able to program is heart.