Create Talking Avatar Videos from Text or Audio
Zero-training lip-sync avatar video generation from audio or text-to-speech, in realistic or cartoon styles, with a free trial token.
Why it matters
Generate lip-sync videos for digital avatars using text-to-speech or audio input. Supports photorealistic and cartoon styles, enabling dynamic avatar communication without pre-training.
Outcomes
What it gets done
Animate digital avatars to lip-sync with provided audio.
Generate talking avatar videos from text input using TTS.
Create digital avatars from images or videos.
Integrate with MCP for seamless workflow automation.
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/vb-flyworks-avatar | bash Capabilities
Tools your agent gets
Create lip-sync video with audio input, animating a digital avatar to match speech in sync.
Create lip-sync video with text input, generating audio from text and animating avatar.
Overview
Flyworks Avatar MCP Server
This MCP server generates lip-synced digital avatar videos from audio or text-to-speech, in realistic or cartoon styles, without requiring model training. Use it for quick avatar-video generation from audio or text. The free trial token has a limited quota, watermark, and 45-second cap; generated video URLs expire quickly and must be downloaded right away.
What it does
Flyworks Avatar MCP generates lip-synced digital avatar videos without requiring any prior model training, from either an audio file or plain text (converted via text-to-speech). It supports both realistic and cartoon avatar styles, and can create digital avatars directly from a source image or video, with both asynchronous and synchronous operation modes.
When to use - and when NOT to
Use this when you want a talking-avatar video quickly - narrating a script with a synced avatar mouth, turning text into a spoken, animated presentation, or creating a custom avatar from your own photo or video clip. The free trial token (2aeda3bcefac46a3) comes with a limited daily quota, watermarked output, and a 45-second duration cap - useful for testing but not production use without your own paid token. Video-based avatar creation produces better quality than image-based creation but takes longer to process. In synchronous mode, the maximum wait time is 10 minutes; generated video URLs are temporary and must be downloaded promptly before they expire.
Capabilities
create_lipsync_video_by_audio- creates a lip-synced avatar video from an audio inputcreate_lipsync_video_by_text- creates a lip-synced avatar video from text, generating the speech audio via text-to-speech first
How to install
Via Smithery:
npx -y @smithery/cli install @Flyworks-AI/flyworks-mcp --client claude
Claude Desktop configuration:
{
"mcpServers": {
"flyworks": {
"command": "uvx",
"args": ["flyworks-mcp", "-y"],
"env": {
"FLYWORKS_API_TOKEN": "your_api_token_here",
"FLYWORKS_API_BASE_URL": "https://hfw-api.hifly.cc/api/v2/hifly",
"FLYWORKS_MCP_BASE_PATH": "/path/to/your/output/directory"
}
}
}
}
Required: FLYWORKS_API_TOKEN. Optional: FLYWORKS_API_BASE_URL, FLYWORKS_MCP_BASE_PATH.
Who it's for
Content creators and developers who want quick, no-training-required talking-avatar videos from audio or text, in realistic or cartoon styles, for demos, presentations, or prototyping before committing to a paid production token.
Source README
An MCP server that provides free and fast lip-sync video creation for digital avatars without pre-training, supporting both photorealistic and cartoon styles with text-to-speech and audio input capabilities.
Installation
Smithery
npx -y @smithery/cli install @Flyworks-AI/flyworks-mcp --client claude
From Source
git clone https://github.com/yourusername/flyworks-mcp.git
cd flyworks-mcp
pip install httpx "mcp[cli]>=1.6.0"
Using uv
uv pip install httpx "mcp[cli]>=1.6.0"
With All Dependencies
pip install pygments pydantic-core httpx "mcp[cli]>=1.6.0"
Configuration
Claude Desktop
{
"mcpServers": {
"flyworks": {
"command": "uvx",
"args": [
"flyworks-mcp",
"-y"
],
"env": {
"FLYWORKS_API_TOKEN": "your_api_token_here",
"FLYWORKS_API_BASE_URL": "https://hfw-api.hifly.cc/api/v2/hifly",
"FLYWORKS_MCP_BASE_PATH": "/path/to/your/output/directory"
}
}
}
}
Available Tools
| Tool | Description |
|---|---|
create_lipsync_video_by_audio |
Create lip-sync video with audio input, animating a digital avatar to speak in sync... |
create_lipsync_video_by_text |
Create lip-sync video with text input, generating audio from text and animating... |
Features
- Create lip-sync videos using digital avatar video and audio as input
- Create lip-sync videos from text (with text-to-speech)
- Generate digital avatars from images or video
- Support for both asynchronous and synchronous modes
- Support for photorealistic and cartoon avatar styles
- Free trial access with limited quota
Environment Variables
Required
FLYWORKS_API_TOKEN- API token for authenticating with the Flyworks service
Optional
FLYWORKS_API_BASE_URL- Base URL for the Flyworks APIFLYWORKS_MCP_BASE_PATH- Path to the directory where output files will be saved
Usage Examples
Create a lip-sync video using an avatar and audio file
Generate a speaking avatar video from text using text-to-speech
Create a digital avatar from an image or video
Convert text to speech and animate an avatar to pronounce it
Resources
Notes
A free trial token '2aeda3bcefac46a3' is available with limited daily quota, watermarked videos, and a 45-second duration limit. Creating avatars from video provides better quality but takes longer than image-based creation. Maximum wait time for synchronous mode is 10 minutes. Video URLs are temporary and should be downloaded immediately.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.