Media Prompt Builder

HeyGen Avatar Prompt Builder — Synthesia & D-ID

Name: HeyGen Avatar Prompt Builder — Synthesia & D-ID
Brand: Media Prompt Builder
Price: 9.99 USD
Availability: InStock

$9.99 USD

Sale Sold out

Taxes included. Shipping calculated at checkout.

Everyone can tell when an avatar video is fake, and almost nobody can say why. It is rarely resolution. It is stillness — a face that does not move between words, blinking on a metronome, hands gesturing against the beat. Fix those and the same tool produces something people actually watch.

This builder writes the brief that fixes them. It is a Google Sheets template — open it, click Make a copy, and it is yours forever. No subscription, no video background needed.

How it works

Open your copy of the builder
Fill Column D using the dropdown menus — 32 fields across 6 sections
Paste your script and describe the presenter you want
Five complete outputs appear instantly on the right
Paste into HeyGen, Synthesia, D-ID or any avatar platform

The 5 outputs you get every time

Main Avatar Prompt — full structured brief covering presenter, voice, framing and output
Avatar Appearance Brief — wardrobe, expression, gestures and the rules that avoid the uncanny valley
Voice and Lip Sync Brief — voice character, accent, pacing, and exactly where lip sync breaks
Framing and Set Brief — eye line, headroom, background depth and safe areas for captions
Production and Consent Plan — batch localisation, disclosure rules and permission requirements

The uncanny valley is solvable

Micro movement beats resolution — a perfectly still face reads as dead however sharp it is
Blinking must be irregular — metronomic blinking is the fastest tell
Gestures on the beat — hands moving against the words is worse than no hands at all
Wardrobe must match the setting — a suit in a bedroom breaks the illusion instantly
Blurred backgrounds hide artifacts better than flat colour does

Lip sync is where avatar video actually fails

Drift almost always appears at the end of a long take, so watch the final seconds. Short sentences sync far better than long clauses — break the script at natural breath points. Numbers, acronyms and brand names are the usual mispronunciations, so test them on their own first. And if the voice is right but the mouth is wrong, regenerate the sync rather than the whole video.

One avatar, every market

The real efficiency of this format is not making one video faster. It is making the same video in fifteen languages without a reshoot, then updating the script next quarter while keeping the presenter. There are dedicated fields for localisation planning, batch production and template reuse.

25 use cases covered

Product explainers · course lessons · corporate training · sales pitches · onboarding · FAQ answers · social ads · testimonial style · multilingual localisation · app walkthroughs · recruitment · webinar intros · support answers and more — across HeyGen, Synthesia, D-ID, Runway Act-Two, Hailuo H3, Colossyan and Elai.

Consent is not optional

Never create an avatar of a real person without their written permission. That includes employees, who must consent explicitly rather than by assumption, and it certainly includes public figures. Disclosure rules differ by platform and country and are tightening. A synthetic presenter delivering medical, financial or legal information carries extra duty of care.

There is a consent field, a disclosure field, and a Consent On File column in the log — because a tool that makes synthetic people should make the paperwork habitual.

Also included

Avatar Presets tab — 10 ready-made setups: Corporate Explainer, Social Ad Presenter, Course Instructor, Healthcare Info, Green Screen Asset and more
Avatar Library tab — uncanny valley fixes, lip sync, framing that builds trust, batching and consent
Avatar Log tab — track every video by avatar, tool, language, length and consent on file
Formula Lock — only the input column is editable, so the tool cannot break by accident

An honest note

For high stakes brand work, a real person on camera still usually wins. This format is at its best for volume, localisation and content you would otherwise never film at all.