Mavericks Adds a Real Avatar Feature That Recreates a Person From One Photo to NoLang

Mavericks, Inc., the Japanese company behind the video generation AI NoLang, has released a new real avatar feature that recreates a person from a single photo. The feature turns one still image into an avatar that moves smoothly inside a video. Putting an avatar of a real person into training, IR, and PR videos makes the speaker feel closer to viewers and builds deeper trust.

The real avatar feature recreates a person from a single photo

NoLang is Japan’s first video generation AI service: users enter text, PDF or PPTX documents, a website URL, or image, audio, and video files, and a video is generated in as little as a few seconds. Since its release in July 2024, NoLang has passed 150,000 registered users, and more than 60 companies now use it across Japan (as of November 2025). Until now, avatar-based videos in NoLang were normally made with the 100-plus built-in avatars or with Live2D and PSD assets that a company already owned.

The new real avatar feature changes that. Users upload a single photo of their face, create a faithful digital double of themselves, and animate it directly in NoLang. Anyone can now express themselves on video with little effort. An avatar of yourself speaking to the audience opens up new creative options for social media posts and user-generated content.

In business settings, putting yourself into every video adds movement and variation to the content. That helps knowledge stick in training videos, strengthens engagement in social media ads, and raises the credibility of presentation videos. Because a high-quality video of the person speaking can be generated from a PC or a smartphone, the feature makes the speaker’s individuality clear and gives personal branding through video a strong push.

How the real avatar feature works

The feature is available on the Business plan. This section explains what it does and how to use it.

Create a real avatar that moves in your video from one photo

No 3D scanning or modeling skills are required. Upload one photo taken from the front, and NoLang generates a real avatar that animates naturally inside the video. It works with photos of real people and also with illustrations and character images, so companies can animate their own IP or mascot characters for PR and communications.

Text or documents are all a finished video needs

A real avatar can be used in NoLang video generation right away. From there, all you enter is the information you want to convey, as text or as a PDF or PPTX document. The AI writes the script automatically and sets sound effects and background music. In about a few dozen seconds, NoLang generates a video in which your avatar delivers the presentation fluently. Managers and executives who cannot find time to shoot footage can therefore appear in videos easily.

”My Voice”: your own voice from about one minute of audio

The voice clone feature can also make the avatar sound like the person it depicts. Record about one minute of audio in the browser or upload an existing audio file, and the AI analyzes the qualities and intonation of the voice to create an original clone voice that reads text aloud.

What multilingual support adds in daily operation

Adopting NoLang with the real avatar feature does more than cut shooting work: it makes video operations far more flexible and extends their reach. A real avatar can speak not only Japanese but English and other languages. Even when the person does not speak the foreign language, the AI voice delivers the message with fluent pronunciation. Product PR videos for overseas markets, training videos in the native languages of foreign workers, and timely video disclosures to overseas investors by listed companies all become straightforward, with no interpreting or dubbing work.

Where the feature is used

The real avatar feature is especially effective in business situations where who is speaking matters.

Training and educational videos

When a senior colleague or a well-known in-house instructor explains the material as a real avatar instead of a robotic machine voice, trainees feel closer to the speaker. The visual movement also holds their attention, which lifts motivation and knowledge retention considerably. If the training content has to change, editing the text in the NoLang editor is enough to update the video.

IR, shareholder meetings, and messages from the top

For executives with packed schedules, finding time for a video shoot is a real obstacle. With a real avatar and a clone voice, earnings explanations and announcements of management policy can be turned into video on time without taking any of the executive’s working hours. Generation in English as well as Japanese is supported, so speaking to overseas investors in one’s own image and voice becomes easy.

Internal newsletters and sharing company values

The enthusiasm and nuance of a founder are hard to convey in text and images alone. Delivered through the founder’s own avatar and voice, the message reaches employees far more deeply, which helps build a sense of unity and raises engagement. Research supports this: at companies where leaders send messages by video, the share of employees who feel engaged with their company is about 33% higher than at companies where they do not, and alignment with goals, connection with colleagues, and the feeling of being valued by the company are each around 10 points higher¹.

Communicating continuously through a founder’s avatar and voice can be expected to lift trust and unity among employees in the same way, and to raise engagement across the organization.

What comes next

This update sharply expands what NoLang can express and what it can be used for. Mavericks, Inc. will continue to advance its avatar technology and to develop customizations for the needs of individual companies, supporting innovation in communication built around video.