Synthesia AI Startup Presents AI Avatar Upgrade For Human Emotion Expressions

JAKARTA - Synthesia artificial intelligence (AI) startups, backed by Nvidia, have introduced a new upgrade that allows AI avatars to convey emotions and human movements.

On April 25, the company revealed its "Avatar Expressive", which aims to express emotions based on text instructions for the company's presentation, marketing, and training goals.

Generative AI is often praised for its ability to create realistic moving images, as is OpenAI's Sora video generator.

However, AI is not without shortcomings, especially when it describes humans, which are often displayed with distorted hands or limbs, collage background, or lips inconsistent with speech.

Synthesia wants to improve this in its latest version, developed by humans who actually read scripts in their studio. This is done to help bots capture lip tracking and be more accurate in their emotional expressions.

Victor Ribarbelli, co-founder and CEO of Synthesia, said in a video that the missing cut was that, unlike humans, "avatars don't understand what they say," which previously led to a lack of facial response to emotions.

In the studio, they were trained to respond to simple instructions like "I'm happy. I'm sad. I'm frustrated" by conveying facial expressions and tones that match the emotions given.

The new avatar is also available in over 130 languages and can provide their own closed text and even clone the user's voice.

However, from model avatar examples on the Synthesia website that speak in languages other than English such as French, German, and Spanish English models are the most advanced and human-like.

The startup is reported to have at least half of Fortune 100 companies listed as clients and provided services to more than 55,000 companies. This includes leaders in various industries such as Zoom, Xerox, Microsoft, and Reuters.

Synthesia is a British-based company founded in 2017. Thanks to the AI boom in the past year, the company has reached a valuation of nearly $1 billion with key support such as Nvidia, which currently dominates the development of AI semiconductor chips.

Because of its narrower approach to creating human-like avatars for the use of Synthesia's business has avoided part of the fierce excitement and competition between rival chatbot models such as OpenAI's ChatGPT and Google's Gemini chatbot.