Talking-avatar API for B2B products
One clip in.
Every pitch out.
Give your users a video twin. Send text, get back a lipsynced video of a real person saying it — personalized for every lead, at API speed.
curl https://pitchtwin.ai/api/v1/generations \
-H "Authorization: Bearer $PITCHTWIN_KEY" \
-d '{
"avatar_id": "av_8f2c",
"script": [
{ "text": "Hi Sarah, saw Acme just raised —" },
{ "text": "congrats!", "pause_after_ms": 300 }
],
"webhook_url": "https://you.app/hooks/pitchtwin"
}'Real footage, not a render
We lipsync over a clip of the actual person. Identity is exact by construction — no uncanny re-generated face.
Built to be embedded
Two endpoints and webhooks. Send plain text, or a script with pauses where you want them.
Priced for volume
Pay per second of video, from under $1 a minute. No seats, no plans, no watermark.
How it works
- 1
Record once
A 30-second clip of the person talking to camera. That's the avatar.
- 2
Send text
Your product calls our API with the script — per lead, per user, per event.
- 3
Ship the video
We call your webhook with a finished MP4 of them saying it. Failed renders are refunded.
Prepaid tokens
1 token = 1 second of finished video. Tokens don't expire. Top up whenever.