AI Makes Photos Talk

And now – you can not only animate your still photos with AI, making the faces in them move slightly, you can make them talk.


Here is an example using my still photo, where my lips are moving. The voice is my cloned voice, it doesn’t sound exactly like mine, simply because fewer people in the world sound exactly like me compared to the other voices, so the AI had fewer samples to learn from to clone my voice exactly across all the words in the English language. But it’ll do for this example.


So i just uploaded a single photo, together with an audio file of my voice saying something. Or I can just use another voice that I like, and have it say what I have typed.


Then I pressed a button, and within seconds I have a talking photo, complete with some slight side-to-side head movements and perfectly-timed blinking eyes.

It was never possible to do this so quickly and easily before.

The service that I used to animate my still photo is the best among similar services that I’ve tried.


Sure, it still looks a little strange. But it’ll get better with time.


How can this tech be perfected?


By moving the body (especially the shoulders) ever so slightly as well, moving the head slightly forward and backward in addition to the slight side-to-side movements, and better synchronisation of the mouth with the speech, to suit the tone of the voice.


Now even if it’s perfected, you’re probably not going to use it this tech for longer videos.


Because if you look at or listen to it longer than a few seconds, your brain will just know I’m not really moving because of how still the rest of the image is, and that the animation doesn’t match the tone of my voice, even if my mouth is synchronised with my speech.


So, for this to work, there must be a higher-level tech that completely mimics you, your speech and your complete whole body movements (for near and far shots, and even side shots), and how your face, intonation of voice and body movements would “normally” behave depending on the content of your speech, including how excited, calm or indifferent you are, on top of your usual speaking style.


This is why when you see an animated photo-realistic avatar speaking, your brain somehow knows it’s not that person’s voice even if it’s cloned, nor is that person really moving even if they somehow look perfect— and the illusion breaks.


Your brain takes in all the cues you are not even aware are there, to form your experience of a real person speaking. This experience remains stored in your memory. So a computer-generated person that is created without all the nuances and subtleties stated here continues to be unreal when looked at or listened to for more than a few seconds.


Nevertheless, can you apply this tech to create or multiply your wealth — completely passively?


Well, when it’s perfected, trainers or thought leaders can say whatever they want on camera — without actually having to do anything.


However, it will be too much to use a completely still photo that is animated this way to deliver a presentation that is an hour or two long for the reasons above.


Nevertheless, it can be one of the spokes in a larger wheel.


An AI can generate a script for a topic, among a list of topics — also generated by an AI.


Another AI looks through the list and decides which topic to work on first.


Another AI creates a video script based on that topic.


Another AI creates the voiceover.


Another AI creates a video with the voiceover and images matching the script.


Another AI writes the title and description of the video.


Another AI uploads the finished video automatically to YouTube for distribution.


And another AI creates short snippets of the AI-generated long-form video for short-form video platforms like TikTok, Instagram or YouTube Shorts, and uploads it to all of them.


Sounds like a dream?


It’s not.


There are already AIs doing all of the above parts, but separately.


And there are surely teams of people working on integrating all those separate components together to provide an All-In-One solution from start to finish.


How valuable would such a tool be?


The time savings it would create would be significant, not to mention the savings in effort and money in not having to pay for each AI separately.


In the meantime, we now have a variety of separate AIs to use and choose from, even if the All-In-One solution is going to take some time to come.


AI is a Web3 concept, being used to enhance Web2 concepts in the form of the production of content for social media platforms for marketing and publicity purposes to generate income for us.


If you know how to use technology to the fullest — you will always have a huge advantage over those who don’t.

Soon to come — how AIs can create full-body animation that replaces our physical body completely and the viewer will have no clue it’s not really us. Watch out for it.