Twinning with AI: Create your digital twin

Express News Service

Create your AI avatar. Synthesise your voice. Animate them. And, voila, you have an AI twin. Easy to say, but is it really that easy to execute? As boundaries blur between the real and virtual worlds, deep tech is making it possible to go beyond creating characters in 2D.

Platforms like Synthesia, Midjourney, DALL.E 2 and Lensa are the go-to places to create your digital twin. The technology can be used as an introductory video of yourself when you send in a resume or are meeting clients online.

Get, Set, Go
While most platforms charge between $10 and $40 a month, there are some that do a decent job for free, but you may have to compromise on the quality of the final output. The first thing is to upload the right image. Look for one where you are facing the camera—with your entire face visible and not at an angle. This is essential because if the face isn’t fully captured, it would not animate smoothly. Some platforms also ask for 10 or more images, ideally from all angles.

The Right Pitch
Using a feature like Descript’s Overdub, ReadLoud, ElevenLabs or Intelligent Speaker, create 
a synthetic version of your voice that can read any text given to it. In order to give a perfect ‘voice model’, you need to record at least 10 minutes of audio—some apps also recommend recording 30 minutes or more. Once the file is uploaded, the app takes anything between 12 and 24 hours to generate the synthetic version. Once done, you can use ChatGPT to generate short scripts to be read. In case you don’t want to use a synthetic version, you can record in your voice and upload the same to be used by the digital twin.

Patch It
The last, and the most important, step is to merge the AI avatar and voice, and animate them. Use a tool like D-ID. It is easy to use and free. You need to upload the AI avatar and the audio recording to generate a video.

AI FOR YOUR PETS
If you have an AI twin, can your pet be far behind? In case of your furry friend, the first roadblock is that it would be nigh impossible to synthesise the voice. Nevertheless, with help from apps such as Picsart, you can see your four-pawed kids in an AI avatar. From astronaut style to royal style, your cats and dogs can now have an enviable avatar. All you need to do is upload their 10-20 images and select a ‘style’. 

<br /> Create your AI avatar. Synthesise your voice. Animate them. And, voila, you have an AI twin. Easy to say, but is it really that easy to execute? As boundaries blur between the real and virtual worlds, deep tech is making it possible to go beyond creating characters in 2D.</p> <p>Platforms like Synthesia, Midjourney, DALL.E 2 and Lensa are the go-to places to create your digital twin. The technology can be used as an introductory video of yourself when you send in a resume or are meeting clients online.</p> <!-- WP QUADS Content Ad Plugin v. 2.0.84 --> <div class="quads-location quads-ad241053 " id="quads-ad241053" style="float:none;margin:0px 3px 3px 3px;padding:0px 0px 0px 0px;" data-lazydelay="0"> <!-- WP QUADS - Quick AdSense Reloaded v.2.0.84 Content AdSense async --> <ins class="adsbygoogle" style="display:block;" data-ad-format="auto" data-ad-client="ca-pub-7550963679842137" data-ad-slot="3250057321"></ins> <script> (adsbygoogle = window.adsbygoogle || []).push({});</script> <!-- end WP QUADS --> </div> <p>Get, Set, Go<br /> While most platforms charge between $10 and $40 a month, there are some that do a decent job for free, but you may have to compromise on the quality of the final output. The first thing is to upload the right image. Look for one where you are facing the camera—with your entire face visible and not at an angle. This is essential because if the face isn’t fully captured, it would not animate smoothly. Some platforms also ask for 10 or more images, ideally from all angles.googletag.cmd.push(function() {googletag.display(‘div-gpt-ad-8052921-2’); });</p> <p>The Right Pitch<br /> Using a feature like Descript’s Overdub, ReadLoud, ElevenLabs or Intelligent Speaker, create <br /> a synthetic version of your voice that can read any text given to it. In order to give a perfect ‘voice model’, you need to record at least 10 minutes of audio—some apps also recommend recording 30 minutes or more. Once the file is uploaded, the app takes anything between 12 and 24 hours to generate the synthetic version. Once done, you can use ChatGPT to generate short scripts to be read. In case you don’t want to use a synthetic version, you can record in your voice and upload the same to be used by the digital twin.</p> <!-- WP QUADS Content Ad Plugin v. 2.0.84 --> <div class="quads-location quads-ad241053 " id="quads-ad241053" style="float:none;margin:0px 3px 3px 3px;padding:0px 0px 0px 0px;" data-lazydelay="0"> <!-- WP QUADS - Quick AdSense Reloaded v.2.0.84 Content AdSense async --> <ins class="adsbygoogle" style="display:block;" data-ad-format="auto" data-ad-client="ca-pub-7550963679842137" data-ad-slot="3250057321"></ins> <script> (adsbygoogle = window.adsbygoogle || []).push({});</script> <!-- end WP QUADS --> </div> <p>Patch It<br /> The last, and the most important, step is to merge the AI avatar and voice, and animate them. Use a tool like D-ID. It is easy to use and free. You need to upload the AI avatar and the audio recording to generate a video.</p> <p>AI FOR YOUR PETS<br /> If you have an AI twin, can your pet be far behind? In case of your furry friend, the first roadblock is that it would be nigh impossible to synthesise the voice. Nevertheless, with help from apps such as Picsart, you can see your four-pawed kids in an AI avatar. From astronaut style to royal style, your cats and dogs can now have an enviable avatar. All you need to do is upload their 10-20 images and select a ‘style’. <br />