hedra/hedra
MCP ServerHedra Character-3 MCP Server - image-to-talking-character video
check: npx https://www.hedra.com/docs
Hedra Character-3 MCP server. Image + audio → expressive talking-character video. Best in class for stylized character avatars (illustrated, anime, mascots) speaking a script.
MCP Server
- Package:
mcp-server-hedra(community) - Transport: stdio
- Command:
npx -y mcp-server-hedra
Authentication
| Variable | Required | Description |
|---|---|---|
HEDRA_API_KEY | Yes | API key from hedra.com → API Keys |
Available Tools
| Tool | Description |
|---|---|
generate_video | Image/asset + audio → talking-character video |
generate_video_with_tts | Image + text (auto-TTS) → character video |
generate_text_to_speech | Text → speech audio asset |
generate_image | Text → image asset |
create_asset / upload_asset / list_assets | Manage input assets |
list_generations / get_generation_status | Track generation jobs |
list_models / get_credits | Models catalog and account credits |
Configuration Example
[[mcp.servers]]
name = "hedra"
type = "stdio"
command = "npx"
args = ["-y", "mcp-server-hedra"]
timeout_seconds = 120
env = { HEDRA_API_KEY = "{{INPUT:HEDRA_API_KEY}}" }
tools = []Notes: Character-3 handles expressive head/shoulder motion + lipsync from a single still. Use as an alternative to HeyGen when you need illustrated or non-photoreal characters.