JoyPix.ai Launches AI Lip-Sync Video Generation Models 'Motion-2.5' and 'Motion-2.5-Dialog'

JoyPix LLC (headquartered in Koto City, Tokyo) has announced the launch of upgraded versions of its core AI lip-sync generation model 'Motion-2' on its online creative tool 'JoyPix.ai', which provides AI avatars, AI lip-sync, AI image generation, and AI video generation. The new versions, 'Motion-2.5' and 'Motion-2.5-Dialog', are now available.

The Popular AI Lip-Sync Video Model from JoyPix.ai, Now Evolved: 'Motion-2.5'

'Motion-2.5' is a single-character model that generates natural lip-sync videos from a single still image and either text or audio input, applicable to people, animals, and characters. 'Motion-2.5-Dialog' is a two-way dialogue model designed for conversation scenes between two characters, suitable for AI avatar videos, interactive content, short social media videos, and virtual presenters.

・Motion-2.5: https://www.joypix.ai/app/talking-avatar/

・Motion-2.5-Dialog: https://www.joypix.ai/app/ja/ai-avatar-dialogue/

Additionally, 'Motion-2.5' and 'Motion-2.5-Dialog' are available not only via the JoyPix.ai web service but also through API access. API access is provided for companies and developers wishing to integrate JoyPix Motion-2.5 into their own services or existing production workflows.

・Motion-2.5 API Documentation: https://www.joypix.ai/openapi/motion-2.5/

The Acclaimed 'Motion-2' Evolves into the AI Lip-Sync Video Generation Model 'Motion-2.5'

JoyPix.ai's AI lip-sync feature 'Motion-2' has been widely used by general creators as well as enterprise users in marketing, advertising, social media management, and content production. Its ease of use—converting a single image and text or audio into a natural talking video of people, animals, or characters—and high practicality as an AI lip-sync video tool have been highly praised.

The newly launched 'Motion-2.5' and 'Motion-2.5-Dialog' are enhanced versions of the acclaimed 'Motion-2'. They improve critical aspects of AI lip-sync, AI mouth animation, and AI avatar video generation, including audio-to-mouth synchronization accuracy, natural movement of the face, head, and body, and strong preservation of the original image's identity.

'Motion-2.5' supports AI lip-sync video generation for a single person, applicable to talking portraits, singing avatars, virtual presenters, and short videos for social media. Meanwhile, 'Motion-2.5-Dialog' supports the generation of dialogue-based lip-sync videos between two people or characters, meeting broader video production needs such as AI conversation videos, character dialogues, and interactions between digital humans.

AI Lip-Sync Video Generation Demo

Demo videos showcasing AI lip-sync video generation using 'Motion-2.5' and 'Motion-2.5-Dialog' are now available. Through Japanese and English demos, users can observe natural mouth movements synchronized with audio, facial expressions, head and body motions, and conversational expressions between two characters.

・Japanese Demo Video:

・English Demo Video:

Key Features of 'Motion-2.5' and 'Motion-2.5-Dialog'

1. Higher Precision Lip-Sync

Finer synchronization of mouth movements with input audio. Achieves more natural lip movements and expressions not only in regular speech but also in singing content.

2. Natural, Coordinated Movements from Face to Head to Body

Emphasizes natural consistency across mouth, head movements, facial expressions, and body posture. Reduces the unnatural stiffness common in AI lip-sync videos, enabling more vivid visual expression.

3. Strong Preservation of Original Image Identity

Better maintains facial features, character traits, and overall image impression across all generated frames. Enables stable AI lip-sync video production across diverse materials such as portraits, brand characters, animal images, and illustrations.

4. Avatar Video Generation from a Single Image

Enables creation of talking or singing avatar videos without filming or recording—just a single still image and text or audio input. Contributes to efficiency in content creation, social media operations, advertising creatives, and promotional videos.

5. Support for Solo Videos and Two-Person Dialogue Videos

'Motion-2.5' generates AI lip-sync videos of a single character speaking. 'Motion-2.5-Dialog' further supports the generation of dialogue-style lip-sync videos where two people or characters converse. This expands creative possibilities for explanatory videos, interview-style content, character conversations, and interactions between digital humans.

Expected Use Cases

Creating Talking Portrait Videos from Still Images

Convert profile photos, illustrations, and character images into naturally speaking videos. Useful for social media posts, self-introduction videos, and message videos.

Presentation and Explainer Videos Using AI Avatars

Create company introductions, service overviews, tutorials, and intro videos in a virtual presenter format. Produce clear and engaging video content while reducing filming costs.

Singing Avatar Videos

Generate videos where still images of people or characters appear to sing along with voice or music tracks. Ideal for music content, character-based content, and fan videos.

Short Videos for Social Media

Efficiently produce short-form videos using AI avatars or characters for platforms like TikTok, YouTube Shorts, Instagram Reels, and X.

Personalized Spokesperson Videos

Create spokesperson videos for customer messages, campaign videos, advertising creatives, and educational content—without the need for filming.

About JoyPix.ai

JoyPix.ai is a creative platform that enables online access to AI digital humans, AI lip-sync, AI image generation, and AI video generation.

It offers various AI creative functions—such as turning images into videos, making characters talk, and generating dialogue-based avatar videos—within a single workflow.

Catering to creators, marketers, corporate content producers, social media managers, and users in education and entertainment, JoyPix.ai aims to provide an intuitive environment where anyone can create high-quality AI videos.

Future Outlook

JoyPix will continue to advance