Apply a Glossary

Pass a brand_glossary_id on HeyGen videos, templates, Video Agent sessions, batches, and video translations so your terms are spoken and translated your way.

Pass brand_glossary_id at the top level of the request, one glossary per call:

curl -X POST "https://api.heygen.com/v3/videos" \
  -H "X-Api-Key: $HEYGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "avatar",
    "brand_glossary_id": "8c2d5e91a47b4f3c8d1e6a9b2f5c7d40",
    "avatar_id": "YOUR_AVATAR_ID",
    "script": "HeyGen shipped this in 2024.",
    "voice_id": "YOUR_VOICE_ID"
  }'

Where	What the glossary does
POST /v3/videos with "type" of studio, avatar, or image	Pronunciations wherever audio is synthesized from a script
POST /v3/videos/batches	Same, set per item
POST /v3/templates/{template_id}	Pronunciations in the template’s speech
POST /v3/video-agents	Pronunciations in the narration, set once per session alongside an optional brand kit
POST /v3/video-translations and batches	Translation rules in the translated script, set per item in a batch
POST /v3/video-translations/proofreads	Translation rules in the editable subtitles

On Brand covers how the glossary and brand kit divide the work. Scenes with their own audio_url or audio_asset_id play that audio as supplied, and captions keep the original script wording.

An unknown brand_glossary_id fails the request with 400 before rendering, so no credits are spent. Glossaries are workspace-scoped: an ID from another workspace is rejected, so use an API key from the workspace that owns the glossary.