{
 "slug": "voice-first",
 "name": "Voice first, then the face",
 "category": "sound",
 "group": "Voice",
 "media": [
  "video"
 ],
 "checked": "2026-11-13",
 "effect": "The mouth is generated for the voice that ships, not for one that is replaced later.",
 "use_when": "Generated people who speak: testimonials, ads, characters.",
 "phrase": "generate the final voice line first, with its delivery, then give that audio to the video model as the audio reference for the face",
 "pitfalls": [
  {
   "text": "Not yet run by us. The order comes from a vendor launch whose film is a showreel of chosen takes; no failure rate is given.",
   "evidence": "practitioner",
   "source": "Director Chair notes on Chimera (ElevenLabs, Eleven v4 with Seedance 2.5)"
  },
  {
   "text": "A voice reference used directly in the video model is often inconsistent between takes; the vendor's post and our own skill agree on that.",
   "evidence": "practitioner",
   "source": "Director Chair notes on Chimera (ElevenLabs, Eleven v4 with Seedance 2.5)"
  },
  {
   "text": "A cloned voice speaking another language than its source loses its accent by design. Test before relying on an accent.",
   "evidence": "practitioner",
   "source": "Director Chair notes on Chimera (ElevenLabs, Eleven v4 with Seedance 2.5) (the vendor's own documentation, read 22 Oct 2026)"
  },
  {
   "text": "Use your own voice or one with written consent.",
   "evidence": "practitioner",
   "source": "Director Chair notes on Chimera (ElevenLabs, Eleven v4 with Seedance 2.5)"
  }
 ],
 "per_model": [],
 "external_refs": [],
 "licensed": [],
 "site_names": [],
 "related": [
  "voice-line",
  "guide-seedance-2-5",
  "voice-written-to-length"
 ],
 "rights": "own-work",
 "evidence": "practitioner",
 "source": {
  "url": null,
  "note": "Own words from Director Chair notes on Chimera (ElevenLabs, Eleven v4 with Seedance 2.5)"
 },
 "body": "",
 "shown_with": null,
 "url": "https://directorchair.ai/sound/voice-first",
 "markdown": "https://directorchair.ai/sound/voice-first.md",
 "prompts": {
  "chat": "Give me a sound brief: what to record or pick, where it sits on the timeline, the levels, and what to check before export, for a short video about [your product], for [your audience]. The technique is voice first, then the face: the mouth is generated for the voice that ships, not for one that is replaced later. How it works: generate the final voice line first, with its delivery, then give that audio to the video model as the audio reference for the face. It is right for: generated people who speak: testimonials, ads, characters. Not yet run by us. A voice reference used directly in the video model is often inconsistent between takes; the vendor's post and our own skill agree on that. Be concrete: numbers, seconds and settings, no theory. Ask me one question first if something you need is missing."
 },
 "licence": {
  "code": "MIT",
  "text": "CC BY 4.0",
  "third_party": "Clips and card texts inside a \"licensed\" block belong to their makers and are shown with permission and credit."
 }
}