DomoAI Enhances Talking Avatar Feature! Unifies Video Production Flow with Emotion and Conversation Capabilities

DomoAI PTE. LTD (Headquarters: Singapore, CEO: Joe Lam, hereinafter 'DomoAI'), which operates the AI video and image generation platform 'DomoAI', has updated and launched its 'Talking Avatar' feature. With this update, voice, emotion, and conversation expressions have been enhanced, enabling the creation of content with storylines and directing in addition to conventional video generation. Furthermore, the entire process from generation to directing and conversation can now be produced as a seamless workflow. Along with this, we provide an integrated environment for producing videos where characters move and speak.

■ Key Updates to the Talking Avatar Feature The Talking Avatar is a feature that generates characters based on images or videos and creates videos of them speaking naturally using only text input. Without any audio recording, you can add voice and facial expressions to a character and generate conversational content.

In this update, centered around this Talking Avatar feature, we enhanced expressive elements such as voice, emotion, and conversation. As a result, the range of character expression in video content has expanded, supporting the creation of content involving storylines and direction.

- Improvement of voice quality (Natural voice model) Achieves more human-like voice expression.

- Addition of emotional expression (Emotion control) Allows setting tones such as 'hope', 'whisper', and 'anger' for each line of dialogue.

- Realization of conversation generation (Multi-voice) Set multiple voices to generate dialogues between characters.

- Smart paste feature (Automatic script splitting) Automatically splits inputted text by roles.

- Support for video uploads (Video material input possible) In addition to portrait images, inputting video materials is also supported.

■ Expansion of Production Flow through Integration with Talking Avatar A major feature of this update is that the entire process—from character generation to adding movement and conversational expressions—can be produced as a continuous workflow.