Do you notice that talking avatars are trending on social media? Many creators are using talking photos to boost their reach on TikTok, YouTube, and Instagram.
D-ID Studio is one of the best talking photo generators, but the problem is that you have to pay to generate videos on D-ID Studio. But don’t worry! In this video, I will walk you through a step-by-step guide to create a talking AI avatar for free and without a watermark. So without further ado, let’s dive into the video.
First, we need to generate a portrait photo to make our talking avatar. You can use any image-generating AI tools of your choice, but in today’s video, I will be using Leonardo AI to generate our image. To get access to Leonardo AI, check the link in the description. Now, register an account with your Gmail. Once done, your dashboard should look like this. Now, click on “Image Generation” and enter your prompt. Don’t worry, I will provide the prompt in the description. You can take your time to modify it or use your own prompt. Now, select a fine-tuned model. I will go with Leonardo Kino XL. After that, select the aspect ratio. I will go with 9:16. After everything is set, hit the “Generate” button. It will take some time to generate the images. Here is our result: Leonardo AI generates four images for us. Now, you have to select the best one. If you are unsatisfied with the result, you can always regenerate it. Once you find your desired image, download it by clicking on the download icon.
Are you enjoying this video so far? If yes, kindly click the subscribe button and turn on the notification bell so that you won’t miss our latest updates.
Now that our avatar image is ready, it’s time to generate a voiceover for our character. If you choose to use your own voice, that’s fine. Otherwise, you can use AI voice generators like ElevenLabs, Fliki AI, Muff, and Lovo AI. Among all, I will be using ElevenLabs. Check the link in the description. After reaching the homepage, log into your account or create a new account with your Google or email account. After successfully creating the account, your interface will look like this. Now, select the voice you wish to use. After that, adjust the voice settings and enter your prompt into this column. After everything is set, click the “Generate” button. Once it is completed, you can preview the voice. If you are satisfied, click the download button to save the audio on your device.
Now that we already have our photo and audio ready, open your browser and search for “TokkingHeads.” Click on the first result from the search bar, and your TokkingHeads interface should look like this. Now, click on “Generate Tokking Heads.” You can create unlimited talking avatars without creating an account. Click on “Upload Your Own Photos” and select the photo you have generated from your device. Once your image has been uploaded successfully, click on the next step. Now you can input audio in three formats: you can upload video clips, audio clips, or generate audio from text. Since you already generated the audio, go to the “Audio” tab and select “Upload Your Own Audio.” There are two options: you can upload an audio file or record yourself. Click here and upload the audio generated by ElevenLabs. After uploading the audio clip, click the next step button. If you want a head avatar only, select the first option. If you want the full avatar, select the second option. After everything is set, click the “Generate Tokking Heads” button. It will take some time to generate our talking avatar.
Now, let me teach you how to remove the watermark. First, open a new tab and log into your Canva account. After reaching the Canva homepage, create a new design with 1080 pixels by 1920 pixels. Upload the avatar image and insert it on the canvas. Adjust the size like this, leaving some space on top of the image. Click on the edit button and crop the image, then select the expand option. Select the size 9:16 and adjust the position to cover the whole frame. Click the magic expand button. It will generate four variants of your image. Select one of them and download it to your device.
Once you have downloaded the image from Canva, go back to TokkingHeads to start the process again. Click here to go to the homepage, then click here to remove your existing photo and upload the new photo. After that, upload the audio file and generate the talking video. After generating the talking video, simply download it by clicking here.
We need to enhance the video to solve the video quality issue. To do this, I will use VMake Video Enhancer. Open a new tab on your browser and search for “VMake Video Enhancer.” Click on the first link. After reaching this page, create an account with your Google or email account. Click here and upload your talking head video. It will start enhancing. After uploading the video, once it’s completed, download the video to your device.
You will notice that the watermark is still on the video. Here’s how to remove it for free. Open CapCut, import your talking avatar, and choose the 9:16 ratio. Drag the video to the timeline and adjust the image size to remove the watermark. Once done, click on the export button to download the video.
Alright, our talking avatar is ready! Let’s enjoy it together.
Thanks for watching! Don’t forget to subscribe to this channel. Watch this video to learn how to create VIRAL motivational videos for MILLIONS of views.
Be the first to comment