Again, Heygen has updated its AI Avatar feature (and there's more to come soon… just keep waiting!), but we'll focus on two major updates: Heygen Avatar IV and product placement for influencers and marketers.
Start with simple product images and scripts and create quick pitch videos to stock influencers while keeping the product for use on social media. Here are the steps required to create a simple video directly from Heygen, as well as some alternative workflows to get high quality results with a few steps and other software tools.
Learn for free!
We offer a free online workshop that will demonstrate these workflows on September 18th. Sign up now! https://luma.com/mcv4um0u
Heygen product placement
Option 1: Stock Talents and Audio
This is a stock method to create a quick “influencer style” product video that introduces handheld products.
Start from the Heygen homepage and select the “Product Placement” tool
This will allow you to select an existing product using sample photos from your library or upload your own samples.
I just looked for a can of “something” online (and wrote a script around it in ChatGpt) and placed the images in the tool.
I chose a stock model from their library.
A few times later I was presented with four different blending options to choose from…but please note that the text on the can is distorted and unreadable when zoomed in.
I then entered the script text I had modified to highlight different words and phrasing.
I was able to choose good voice through the ElevenLabs API, but voice control is really lacking and is very robotic in its raw form. Here's the video rendered straight from Hagen:
https://www.youtube.com/watch?v=gobqfnkertc
Option 2: Stock Talent and Eleven Love Audio
This is my favorite option. Because you can control the response of your voice and talent in order to become more natural based on your voice and voice inflection and energy.
The workflow is the same until you reach the Heygen VO script. Next, upload the pre-recorded audio tracks. Here I used 11 voices that worked well on Hagen's stock model
And here's the video of the result. This time we'll choose a different angle provided by Heygen.
https://www.youtube.com/watch?v=-yl1mvi_ts
Option 3: Generates only product images
This workflow allows you to use photos of anyone or AI-generated talent and products to generate only images of talent holding the product. You can change this in Photoshop or Autpaint in Midjourney and then go back to Heygen to add animations and audio.
I started with a model generated in Midjourney and tried to clean up the images in Photoshop and use them in the Heygen product placement tool.
Unfortunately, my model is completely distorted, and the result of Deafulat is someone's living room or bedroom background, and there's no way to change it. So this time I went in a different direction.
So I had to find another way to get the product from her and take this a different approach.
After some attempts with other tools, including Adobe Firefly in Photoshop, I discovered that Krea AI gives me what I need.
I took the edited image to Heygen using the Video Avatar IV tool using the Image and used the same audio file I created previously in ElevenLabs with some simple instructions. The results were exactly what I was looking for!
Here's the video completed directly from Heygen:
https://www.youtube.com/watch?v=bjq-xpjmiwc
As you can see, I now use a green screen background to greet my talent precisely and put her everywhere and in settings.
Use Video Avatar IV from Photos
This process continues where you left off using the Option 3 workflow above.
Select a photo to your photo using the Avatar IV option from the Heygen homepage and upload an audio recorded from ElevenLabs and an SFX edited image.
I was tired of the Bloody Mary Mix can so I tried another project using another humorous script supported by Power Tool and ChatGpt. This took me a lot longer than I originally thought, as this gave me everything except what I wanted by getting the prompt to work properly for a shot of the Broll! lol
I was chatgpt with Assist for script writing and then went to ElevenLabs to record 11 audio and SFX.
I created the model in Midjourney, brought it to Heygen with a photo of the drill, combined it with the same image using the product placement tool, then cleaned it into Photoshop and replaced the background with Greenscreen.
I saved the best images from the product arrangement, took it to Photoshop, cleaned it and replaced the background.
I started by using the generation extension to get the framing right.
I placed a solid green screen layer underneath it and the combined images were what I needed for my avatar.
I put everything back to Heygen, used the photos as a video avatar IV tool, uploaded the cleaned images, and generated the main video using VO recordings from ElevenLabs.
I ran a B-roll pass once using only text prompts, but failed miserably, so I created 11 tracks, “Hmmmmmmm” and “Ahhhhhhh” to get the model to work and react. I got some available takes, but most of the time I used other tools to generate most of the Brolls (when it actually worked!)
All other B-rolls were generated with various tools such as Midjourney, Runway, Hailou 2, and graphics were created with Gemini 2.5 Flash.
Video editing to take you to the next level…
Of course, if you have video editing skills, you can shoot all the AI-generated content for your project and edit the finished video using the content you generated yourself.
Everything was traditionally edited in Adobe Premiere Pro, and edited music tracks created with Suno AI.
Note: I know this is a humor of sexist objectification in a rather low Brow…but we all know that sex usually sells products for men (think a tool calendar?). But it's really fun with an extra facilitator! 😉
https://www.youtube.com/watch?v=4ol_syjvrik
Check out Heygen's new Avatar IV tool for yourself and have fun!
