Content
summary Summary

Similar to D-ID's Liveportait tool, Synthesia allows users to create AI avatars that can serve as presenters and TikTok influencers.

Ad

Simplifying content creation

The key feature of D-ID's program is that it allows users to create reels, TikTok videos or entire YouTube episodes delivered by a human-looking AI creator.

Users can enter their own text and choose from over 120 voices and 125 different avatars to tell their stories. B-roll, such as infographics, can also be added along the way.

Synthesia's videos are more stilted and awkward than D-ID's live portrait. Another big difference is that on D-ID's Live Portrait, users can bring to life any avatar they choose, including themselves - long-dead ancestors as part of their partnership with the family search platform My Heritage, or creations users have made themselves on Midjourney or Dall-E.

Ad
Ad

However, Synthesia has more tools to customize and edit the video, making it more suitable for immediate publishing. Background media can be replaced, the video can be cropped, and users can add features such as blur and background music. This streamlines the process of creating video content.

Technology and research

Synthesia's built-in scriptwriting elements are created by GPT 4, while the avatars are created using a technology developed in-house that the team refers to as "text to video".

Synthesia's team consists largely of academics in the field of AI and computer science. They include Matthias Niessner, Professor at the Technical University of Munich, and Lourdes Agapito, Professor of 3D Vision in the Department of Computer Science at University College London (UCL).

Pricing

There is a very limited free demo where users can try out an avatar of their choice. After that, the pricing structure is pay-per-use.

The standard personal plan for $30 gets you 1 seat, 10 credits per month, and a maximum of 10 scenes per video.

Recommendation

There is an enterprise level plan for an unspecified price depending on the individual needs of the company, but it is billed as including unlimited credits.

Content moderation

Moderation of AI content is an important safety issue, especially for public content. All content created on Synthesia is reviewed. The guidelines for content moderation are as follows:

Anyone can open Photoshop, Word, PowerPoint or a game engine like Unreal or Unity and create whatever they want, even the most hateful and vile content.

We take a different approach and have decided to control what content you can and cannot create on Synthesia.

How this "on the rails" moderation and creation process works in practice is unclear, but Synthesia says it's only had to ban a handful of accounts since launching in 2017.

Ad
Ad
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.
Support our independent, free-access reporting. Any contribution helps and secures our future. Support now:
Bank transfer
Summary
  • Synthesia is an AI-powered tool that allows users to create AI avatars as video presenters for reels, TikTok videos, or YouTube episodes using text-to-video technology developed in-house and scripting elements created by GPT-4.
  • The application streamlines the content creation process with various video editing and customization features, including background media replacement, cropping, blurring, and background music.
  • While Synthesia's content moderation policy emphasizes control over the content generated within its platform, the company claims to have banned only a few accounts since its launch in 2017.
Journalist and published fiction author Harry is leveraging AI tools to bring his stories to life in new ways. He is currently working on making the first entirely AI generated movies from his novels and has a serialised story newsletter illustrated by Midjourney.
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.