How do you use your own voice in a generated video?
On Higgsfield: In Higgsfield Audio, open Voice Presets, create a custom voice and record or upload an MP3 or WAV sample of your own voice; after a consent step confirming the voice is yours or permitted, the clone becomes a preset whose tone and pace can be tuned. The cloned voice can then read any script in AI Text to Speech, including a translated script, or voice an explainer, a greeting or a personal message.
Without Higgsfield: For Sora 2 or Veo 3.1 clips, Ruslan generates the clip, extracts its audio, separates the voice and trains his own voice in ElevenLabs.
Axis: replace the generated voice with a cloned one after generation … feed your own voice in before generation
Concept: voice cloning
Answers by date: 17 records in 16 sources, newest first; the answers above are the current ones. The advice was refined (2026), never reversed, in these records.
| date | who | answer | source |
|---|---|---|---|
| 2026-10-24 | Adil | To give a character a recorded voice, Adil makes a video element: select the character generation, create an element, choose video, upload the audio, add a short prompt for how the character should move and speak, and generate. | The Complete AI Filmmaking Masterclass Everyone’s Missing (Full Tutorial) 00:04:07 |
| 2026-09-05 | Adil | For a generated presenter of himself Adil used a short excerpt of his own video, with his face, voice, gestures, framing and studio, and advises a clear reference with steady light, visible face and audible speech. | GPT-6 Astra + Higgsfield MCP Made This ENTIRE Video in One Chat 00:04:16 |
| 2026-08-12 | Higgsfield | Instead of a preset, the avatar can use a clone of one's own voice from a recording of 10 seconds to 3 minutes; Higgsfield advises cloning only a voice that is yours or that you may use. | How To Create a Talking AI Avatar in Claude with Higgsfield MCP (Full Workflow + Prompts) |
| 2026-08-07 | Higgsfield | For voice cloning, Higgsfield says reading the provided sample script tends to give a cleaner clone than freeform speech. | How to Keep the Same AI Voice Across Every Video: Consistent AI Audio Explained |
| 2026-08-07 | Higgsfield | To reproduce one particular real voice, Higgsfield says a voice cloned from a real sample holds up more reliably than matching a preset by ear each time. | How to Keep the Same AI Voice Across Every Video: Consistent AI Audio Explained |
| 2025-11-05 | Higgsfield | For instant voice cloning in Higgsfield Recast, Higgsfield advises uploading an audio excerpt, preferably recorded in a quiet room without music or echo. | AI Face and Character Swap in Video & Photo: PRO Guide from Top Engineer |
| 2025-10-15 | Ruslan | To use your own voice in Sora 2 or Veo 3.1 clips, Ruslan generates the clip, extracts its audio, separates the voice, and trains his own voice in ElevenLabs. | Is Veo 3.1 worth the attention? Honest review and results 01:37:51 |
| 2026-10-22 | Higgsfield | To clone a voice, Higgsfield's steps are: record or upload a short clip, confirm in the consent step that the voice is yours or permitted, let the model learn it and tune tone and pace, then paste a script and download the read as MP3, WAV or OGG. · also said: 2026-03-02 Higgsfield Audio: A New Feature That Speaks Volumes. The Ultimate AI Text-to-Speech, Voice Swap, and Video Translation Tool, 2026-07-09 I Automated This Hidden Faceless Niche Using Higgsfield AI + Claude Fable 5, 2026-07-09 How to Build a $10K/Month Faceless YouTube Channel With Higgsfield AI + Claude Fable 5 (Full Guide), 2026-08-01 How do I use lipsync, voiceover, and aspect ratios?, 2026-08-10 7 Best AI Avatar Generators for Talking Videos in 2026 (Tested and Compared), 2026-10-22 AI Text to Speech, 2026-10-22 AI Video Translator | AI Voice Cloning |