Metaphysic PRO

Hands-on toolkit for controllable AI faces, voices, and performances, built for production
Rating
Your vote:
No screenshots
Notify me upon availability

Skip the guesswork and move straight to results. In Metaphysic PRO, create a project, pick a template for film, games, ads, or social, and import reference photos, plates, and guide audio. Attach licensed datasets or your own sources, then lock permissions so only approved users can render. Tag consent and rights, choose a pre-trained face/voice or kick off training, and set export guardrails. The live preview unifies face, voice, and motion on one timeline for rapid iteration. When ready, publish to a staging folder with versioning and audit logs.

To craft a screen‑ready face performance, open Face Builder. Upload scans or stills, pick a topology, and tune expression ranges with quick takes. Drop the talent into your shot; the performance engine maps expression, gaze, and head pose from video or mocap, with retargeting for exaggeration, smoothing, and lag. Use the expression graph to nudge moments—advance brows two frames, soften plosives, or dampen jaw motion in fast cuts. Batch a take across lenses, and export meshes, textures, and animation curves to Unreal, Unity, or your compositor. Every render carries a forensic watermark and full version history.

For voice, create a library entry, attach consent, and record or import samples. Choose a style—narration, conversational, promo—or build your own. Paste the script, set pronunciations with the lexicon tool, and mark pauses and emphases with simple notation. Switch language or dialect on the fly and preview prosody live; mouth shapes auto‑sync so your rig stays in step. For localization, import SRT, time‑stretch to picture, and output dialogue stems plus room tone. Deliver to DAWs as WAV and to smart‑speaker skills as SSML without reformatting.

Deploy PRO in your cloud or on‑prem. Bring your own data, keep models private, and enforce encryption, watermarking, and content approvals. White‑label the interface and expose only the tools teams need. The dashboard tracks usage, GPU hours, and rights; AI optimization proposes render and model settings by project type. Import or export review packages as PDF, PNG, or HTML, and attach media to briefs so producers and clients comment in context. When it’s time to ship, push assets to your pipeline and archive a signed report.

Review summary

Features

  • Project templates for film, games, ads, and social formats
  • Face Builder with scan/still ingestion, topology presets, and expression tuning
  • Performance engine for expression, gaze, and head‑pose mapping from video or mocap
  • Retargeting controls for exaggeration, smoothing, and latency
  • Multilingual neural TTS with style presets, lexicon, and real‑time prosody preview
  • Auto lip‑sync linking voice to face animation
  • Batch processing and timeline editing for precise adjustments
  • Dataset and consent manager with rights tagging and export guardrails
  • Private model training (bring your own data), cloud or on‑prem deployment
  • Security: encryption, watermarking, approvals, audit logs, and versioning
  • Integrations: Unreal Engine, Unity, Nuke/After Effects, Pro Tools/DAWs, smart‑speaker skills
  • White labeling with role‑based access and branded UI
  • Dashboard analytics and AI‑driven optimization suggestions
  • Import/export review packages (PDF, PNG, HTML) and media attachments in briefs

How It’s Used

  • Film/TV: digital doubles, subtle face fixes, ADR and lip‑synced dubbing
  • Games: NPC performances, localized dialogue, real‑time cutscenes
  • Advertising: rapid talent variants, accent testing, script swaps near deadline
  • Virtual presenters and influencers for social content and livestreams
  • Corporate training and e‑learning localization at scale
  • Accessibility: personalized synthetic voices for assistive solutions
  • Smart‑speaker and IVR voice design with SSML delivery
  • VFX/post: gaze correction, expression polishing, face replacements
  • Podcasts/audiobooks: consistent voice clones with pronunciation control
  • R&D and prototyping: secure sandbox for face/voice experiments with governance

Comments

User

Your vote:

Recent downloads