Talking video and voice
over one API
Send one photo and a script or a recording. Get back a video where the face speaks, from $0.10 a minute. Or send text and get an MP3 for $0.05 a minute. Prepaid credits, no subscription, $1 in API credits with your first key, and a free tier to test with.
curl -X POST https://api.stork.ai/api/v1/talking-photo \
-H "Authorization: Bearer sk_stork_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"image_url": "https://www.stork.ai/samples/talking-photo/portrait.jpg",
"text": "Hi! I am your new assistant. Ask me anything, any time.",
"voice_id": "en-leo",
"tier": "free"
}'Works as pasted with your key: our sample photo, a built-in voice, the free tier.
The APIs
Talking Photo API
One photo plus text or audio in, a lip-synced MP4 out.
$0.20 a minute Standard, $0.40 Pro, at 720p
- Type the script: 28 voices in 13 languages
- Or send your own audio, any length
- 9:16, 16:9, 1:1; 480p to 1080p
- Signed webhooks, safe retries
- Free tier: watermarked, up to 60 s
Read the docs →
Speech API
Text in, an MP3 out.
$0.05 a minute of finished audio
- 28 built-in voices in 13 languages
- Or a voice you added to your account
- Long scripts in one file
- Safe retries
- Free tier: 3 files a day
Read the docs →
Singing Video API
One photo plus a song in, a performance video out.
$12.00 a minute, billed by the second
- The person sings your song, lip-synced
- Filmed from several camera angles
- 15 seconds to 5 minutes
- Signed webhooks, safe retries
- Pay for the length delivered
Read the docs →
A live version, where the photo talks back in real time in your LiveKit room, is in early access.
How you pay
One key, one balance, no plan
Prepaid credits
1 credit is 1 cent. Packs start at $2.00 and never expire. The same balance pays for the API and the website.
Pay for what you get
We hold the estimated price when a job starts, charge for the length we deliver, and return every credit if it fails.
Know before you spend
An estimate endpoint prices any job without making it. Auto-recharge is optional and off by default.
Buy credits at stork.ai/credits.
Node and Python libraries
A few lines instead of the plumbing
The storkai package, on npm and on PyPI, wraps every call on this page. It adds the idempotency key, waits for the job, turns errors into one exception type and checks webhook signatures. The Node package includes TypeScript types; the Python package has no dependencies.
npm install storkai
export STORK_API_KEY=sk_stork_...
import { Stork } from "storkai";
const stork = new Stork();
const video = await stork.talkingPhoto.createAndWait({
image_url: "https://www.stork.ai/samples/talking-photo/portrait.jpg",
text: "Hi! I am your new assistant.",
voice_id: "en-leo",
tier: "free",
});
console.log(video.video_url);pip install storkai
export STORK_API_KEY=sk_stork_...
from storkai import Stork
stork = Stork()
video = stork.talking_photo.create_and_wait(
image_url="https://www.stork.ai/samples/talking-photo/portrait.jpg",
text="Hi! I am your new assistant.",
voice_id="en-leo",
tier="free",
)
print(video["video_url"])Every API has create, get, list, estimate, wait and createAndWait (create_and_wait in Python). The full references are on npm and PyPI.
n8n
A node for n8n workflows
The community node n8n-nodes-storkai puts Talking Photo, Speech and Singing Video in a workflow, each with Create, Estimate, Get and Get Many. Create waits for the finished file by default, so the next node receives the video or audio address.
In a self-hosted n8n, open Settings, then Community Nodes, choose Install and enter n8n-nodes-storkai. Step-by-step guide with a workflow to import. The source and a sample workflow are on GitHub, the package on npm.
For AI agents
The same calls as MCP tools
Add Stork as a remote MCP server and an agent can price, make and fetch a talking video by itself: stork_talking_photo_create, _status, _voices and _estimate. Works with Claude, Cursor, VS Code and any MCP client.
{
"mcpServers": {
"stork": {
"url": "https://mcp.stork.ai/mcp",
"headers": { "Authorization": "Bearer sk_stork_YOUR_KEY" }
}
}
}Machine-readable: openapi.json · llms.txt · llms-full.txt
More
AI tool directory API
Search AI tools and MCP servers
The same MCP server and REST host also search the Stork directory of AI tools and MCP servers, and look up first-party LLM pricing. It is free and needs no key.
curl "https://api.stork.ai/api/v1/tools?q=code+editor&limit=3"{
"mcpServers": {
"stork": {
"url": "https://mcp.stork.ai/mcp"
}
}
}claude mcp add stork \
--url https://mcp.stork.ai/mcpLLM Pricing
stork_llm_pricesFirst-party LLM list prices with the per-call server-side web-search fee billed on top of tokens — plus long-context tiers, cached-input rates, and the timestamp each figure was last verified. No reseller rates.
AI Tool Discovery Tools
stork_tools_searchSearch the directory of AI tools by query, category, pricing model, or taxonomy facet
stork_tools_detailsFull tool details: pricing tiers, features, integrations, competitors, reviews
stork_tools_compareSide-by-side comparison of 2-5 AI tools
stork_tools_recommendSemantic recommendations: "what tool should I use for X?" with confidence scores
stork_tools_alternativesFind competitors and alternatives for any tool
stork_tools_categoriesBrowse the full taxonomy: 9 facets, 64 subcategories, 233 leaves
MCP Server Discovery Tools
stork_searchFind MCP servers by natural language description
stork_server_detailsFull server details with trust scores and tool inventory
stork_get_install_configReady-to-paste config for Cursor, Claude, VS Code, Zed
stork_compareCompare 2-5 MCP servers side by side
stork_list_filtersBrowse categories and hosting filters
REST API
Structured data, ready to use
All endpoints return JSON with CORS enabled. Base URL: https://api.stork.ai
Free: 100 requests a day per IP address. No API key required.
/api/v1/toolsSearch AI tools or get a specific tool by slug.
Parameters
qslugcategorypricingfacetIdlimitcurl "https://api.stork.ai/api/v1/tools?q=writing+assistant&pricing=free&limit=5"
{
"results": [
{
"slug": "grammarly",
"name": "Grammarly",
"description": "AI-powered writing assistant...",
"url": "https://grammarly.com",
"pricing": "freemium",
"category": "Writing",
"tags": ["AI", "writing", "grammar"],
"hasFreeTier": true,
"keyFeatures": ["Grammar check", "Tone detection"],
"storkUrl": "https://www.stork.ai/en/grammarly"
}
],
"total": 1
}/api/v1/tools/recommendSemantic recommendations. Describe your task, get the best tools with explanations.
Parameters
query (required)limitcurl -X POST "https://api.stork.ai/api/v1/tools/recommend" \
-H "Content-Type: application/json" \
-d '{"query": "transcribe meeting recordings", "limit": 3}'
{
"summary": "Tools for meeting transcription and summarization.",
"recommendations": [
{
"slug": "otterpilot",
"name": "OtterPilot",
"reason": "Designed specifically for meeting transcription...",
"confidence": "high",
"storkUrl": "https://www.stork.ai/en/otterpilot"
}
]
}/api/v1/tools/compareCompare 2-5 tools side by side on pricing, features, and reviews.
Parameters
slugs (comma-separated)curl "https://api.stork.ai/api/v1/tools/compare?slugs=cursor-30d34,grammarly"/api/v1/tools/alternativesFind competitors and alternatives for a tool.
Parameters
slug (required)curl "https://api.stork.ai/api/v1/tools/alternatives?slug=grammarly"/api/v1/categoriesList all AI tool categories with tool counts.
curl "https://api.stork.ai/api/v1/categories"/api/v1/taxonomyBrowse the full 3-level taxonomy tree. Optionally filter by facet.
Parameters
facetIdcurl "https://api.stork.ai/api/v1/taxonomy?facetId=build"npm Package
Local / offline use
Install the storkai npm package for local stdio transport.
npm install -g storkai
# Or use npx in your MCP config:
{
"mcpServers": {
"stork": {
"command": "npx",
"args": ["-y", "storkai"]
}
}
}The directory API is free: 100 requests a day per IP address, no key. Keys from the dashboard are for the media APIs above.
