I am inundated with new ways AI can be used in my business and everyday life. I’m constantly trying to keep up with companies on the cutting edge of AI, offering innovations that can help me in various aspects of my life. It’s a firehose of information.
Jeremy Caplan introduced me to the AI video platform Hypernatural. I subscribe to Caplan’s email newsletter, Wonder Tools (you should sign up) , and he introduced this tool in his July 18, 2024, newsletter. It’s just one of many available AI video tools, but his review made me want to try it.
I have been watching AI video creation apps for a while now, and frankly, some of the videos I have seen have been bonkers—characters that morph and disturbingly move around the scene. I get it. We are still in the early stages of AI video development, and what we can do today will be miles away from what we can do next year. But I don’t want to wait, so I thought I might dip my toes into it with my blog posts.
As I have mentioned in other blog posts, I am a big fan of COPE, the “Create Once, Publish Everywhere” methodology popular among content creators. For instance, I used the text as chapters in a book when I wrote episodes of my “Alabama Short Stories” podcast. I recorded for the podcast, and the audio was posted on YouTube. However, YouTube needed some video, so I used a static image and an audiogram, which bounced up and down as I spoke. It’s not great, but I didn’t have the time or resources to create a legitimate video.
If you have been reading my blog posts on shawnwright.net, you will notice that at the bottom of each post, I have recorded an audio post called “From Paste-Up to Pixels.” This podcast is not broadcast to podcast apps because that costs money.
I thought I would turn my latest post, “The Importance of Catching Typos: A Cautionary Tale,” into a video using Hypernatural. You can view the video below. Here is what I learned.
You have to open a free account to get started, which should be sufficient for you to try. The videos you can create have a maximum length, and a watermark is placed in the bottom right corner. I chose to create a “Promo for a Podcast.” I could have linked it to a published podcast, but I just uploaded my final mp3 file.
You are given options for caption styles and custom colors, which businesses will appreciate to stay on brand. I added a logo in the top left corner and chose my video style. There are many different styles, such as anime, fantasy, goth, impressionist, Mickey Mouse, psychedelic, and photographic. Photographic was way too weird because it is not, but it is, photographic. It has that weird AI look. I wanted to use something different, so I chose Cubism.
Hypernatural went right to work, transcribing my audio and creating images related to what I was saying. In any case, there was movement, and it looked like a video. Captions appeared on the screen as the audio played. Now, here is where the problem was for me. It didn’t get all the words or punctuation correct. Maybe that is not a big deal for most, but this article is about catching typos. I spent a lot of time getting spelling and punctuation correct. Now that I have preached about making sure you proof, I am sure there will be a typo here. I can’t see it, but I must let this out in the world.
Am I going to start using AI video? For the odd project here or there, but not every day. Talk to me next year—the quality may be excellent by that time. That is another reason I won’t be using it often. When I look back on this video in a year, I will probably be appalled by the quality and want to update it. I will be updating my videos all the time.
So go check out Hypernatural and see if it will work for you. Here is a video showing what you can do with it.