I created an interactive digital avatar of myself — and you can talk to it
Synthesia built a talking digital twin of a journalist that only answers questions about a single article, showing how AI avatars can be personalized and deterministic.
Why Now
The story follows Synthesia’s recent $4 billion valuation and its launch of interactive avatar products, prompting the author to create a personal avatar for herself.
What Happened
Synthesia used a mix of voice‑to‑text, language, text‑to‑voice, and video models to create an avatar that can talk back and listen. The avatar was trained only on the author’s article about venture‑backed startup fraud and will only answer questions about that story. The creation took a few days and involved photo and voice recordings of the author.
Why It Matters
Deterministic avatars can provide consistent, fact‑bound responses, useful for PR or training scenarios, but they also raise concerns about limited scope and potential misuse. The tech shows how quickly a personalized digital twin can be built, hinting at broader adoption in journalism, corporate training, and customer service.
The Limitation
The avatar’s responses are restricted to one article, so it cannot generalize beyond that content and may not sound fully natural or human‑like.
What You Can Do
Try building a simple deterministic avatar with Synthesia’s API to see how it handles a single topic you care about.
Source
Read original sourceWhy we picked this
Discusses interactive AI avatar and AI clones, core AI content.