LTX-2 19B Lipsync
Model Information
Display Name: LTX-2 19B Lipsync
API Model ID: sync-ai/ltx-2-19b-lipsync
Category: Image To Video
Description: LTX-2 19B Lipsync synthesizes high-quality talking head videos with precise lip synchronization from audio input. Built on a massive 19 billion parameter Diffusion Transformer (DiT) architecture for superior visual quality and natural motion. **Key Features:** - 19B parameter DiT architecture - Audio-driven lip synchronization - Optional reference portrait image - Multi-resolution: 480p, 720p, 1080p - Up to 20 seconds of audio-driven video - Optional text prompt for style control **Capabilities:** - Audio-to-video lip-sync generation - Image-guided talking head synthesis - Text-prompted style direction - Automatic head movements and expressions - Multi-resolution output **Best For:** - High-quality talking head videos - Professional spokesperson content - Audio-driven video content creation - Virtual presenter generation - Dubbing and localization - Social media and marketing content **Input Requirements:** - Audio: Audio file URL (required, max 20 seconds) - Image: Reference portrait (optional) - Prompt: Text style guidance (optional) - Resolution: 480p, 720p or 1080p **Technical Specs:** - Architecture: Diffusion Transformer (DiT) - Parameters: 19 billion - Audio processing: ffprobe-based duration detection - Output: MP4 video **Pricing:** - Per-second billing based on audio duration - Tiered by resolution (480p/720p/1080p) - Duration detected automatically from audio **Limitations:** - Maximum 20 seconds audio duration - Single speaker per generation - Best results with clear audio input
How to Use This Model
To use LTX-2 19B Lipsync via the HInow.ai API, use the model ID: sync-ai/ltx-2-19b-lipsync
API Request Example (Image Generation)
POST https://api.hinow.ai/v1/images
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
{
"model": "sync-ai/ltx-2-19b-lipsync",
"prompt": "Your image description here"
}
Pricing
- total: $0.021
Available Parameters
- audio: Audio URL driving the lipsync (required). Duration detected automatically for billing. Max 20 seconds.
- resolution: Output video resolution. Default: 720p. Affects pricing. (Options: 480p, 720p, 1080p)
- seed: Random seed for reproducibility (optional). Use -1 for random.
Quick Reference
To use this model, set: "model": "sync-ai/ltx-2-19b-lipsync"
Featured: No
Documentation: https://hinow.ai/models/sync-ai/ltx-2-19b-lipsync
API Endpoint: https://api.hinow.ai/v1
