Hedra
AI talking-character video — turn a portrait plus audio into lip-synced video up to 10 minutes long.
About
Hedra's flagship Character-3 model turns a single portrait image and an audio track into a talking or singing character video, up to 10 minutes long at up to 1080p, with strong lip-sync across multiple languages. A newer Omnia model handles short full-body motion scenes. Beyond its own models, Hedra runs a studio and API with a catalog of 110+ third-party generation models (video, image, voices) on one credit balance, plus a Live Avatars real-time streaming option.
Pricing was restructured in late 2026: Basic is $20/month for 2,000 credits, Pro $50, Ultra $100, with monthly credit expiry. Character-3 runs roughly 6 credits per second at 720p, so long-form output burns credits fast. The free tier is watermarked and non-commercial, and per-model credit rates aren't published, so budget carefully before committing to a series.
What it can do
- Character-3 image+audio lip-sync up to 10 min at 1080p
- Omnia short cinematic character scenes with camera control
- 110+ third-party model catalog on one credit balance
- Live Avatars real-time streaming (~sub-100ms latency)
- ElevenLabs voice integration with voice cloning
- Developer API with async job polling
Setup
Sign up at hedra.com — no credit card needed for the free tier. Upload a clear front-facing portrait, add audio (upload, record, or TTS), pick Character-3, and generate. API access requires a paid plan and an API key from your profile page.
Prerequisites
A clear portrait image (512px+ recommended) and clean audio or a script for TTS.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 49; Trending score 78/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.