How to Clone Your Voice With AI (and Use It in Videos)

On VdoBloom, AI voice cloning exists in exactly one place: the “Use my voice” control inside Creator Reel. You record or upload about thirty seconds of your own speech once, and every reel you build afterwards is voiced in your voice while keeping lip-sync. There is no standalone voice-clone tool, and the option only appears when cloning is switched on and only for paid accounts.

Updated October 2026.

Where voice cloning actually lives

Most platforms put voice cloning in a settings page and let you use the result anywhere. VdoBloom does not. The capability is bound to Creator Reel, the tool that turns a topic into a scripted, filmed, captioned short. The form asks the server whether cloning is available before it renders the control at all: if the answer comes back that cloning is off, the button reads Use my voice (unavailable right now) and explains that reels will use the default AI voice instead.

That matters for expectations. If what you want is a cloned voice reading an audiobook, or a cloned voice dropped onto an existing video, this is not that tool. What it is: a way to make the talking-head shorts you publish sound like you rather than like a synthetic presenter.

What the toggle does

Inside Creator Reel, the voice control sits alongside the model and aspect options. It reads Use my voice (clone) with an ON/OFF state. Turning it on opens two ways in:

• Record. The tool asks for roughly thirty seconds from your microphone.
• Upload. You supply a clip you already have, as mp3, wav, m4a, ogg, webm, flac or aac, capped at 25 MB.

Once saved, the form recognises your sample on later visits and labels it Your saved voice, so you do not re-record every time. That is the whole feature: one sample, reused across future reels.

During generation, the voiceover that the reel produces is converted to your voice, and the lip-sync is carried across with it. If the conversion fails for a clip, the tool does not fail silently — it either tells you that a stock voice was used because your voice could not be cloned, or that some clips kept a stock voice while others were converted. Either way you are told, and the reel still builds.

Cloning your voice in Creator Reel, step by step

1. Open Creator Reel. It lives at /dashboard/video-creation/creator-reel/ in the video creation dashboard.
2. Confirm your account can use it. On a free account the control shows a lock and a prompt to upgrade. Voice cloning is a paid-tier feature, and the server discards a voice sample submitted from a free account rather than quietly failing later.
3. Write the topic first. The reel is built from a short description of what you want to talk about, not from a script you paste in one go. The script comes back for you to edit, and writing it is free.
4. Add your identity photo. The reel is filmed around a photo of you, so a clear, front-facing shot does the most work here.
5. Turn on Use my voice. Record about thirty seconds, or upload an existing clip. Watch the rights note next to the control: use your own voice, or one you have written permission to use.
6. Review the script, then the clips. Creator Reel shows you the script, then the draft clips, then the captions, and you approve each stage. The cost is charged once the videos generate, not before.
7. Build and download. The final build joins the clips and captions into finished files. A 9:16 reel produces two files, a clean cut and a captioned cut; a 16:9 reel produces four, including vertical versions.

Recording a sample that clones well

• Thirty seconds of continuous speech. A single unbroken take is better than six short clips stitched together.
• Quiet room, steady distance. No music, no fan, no traffic. Hold the microphone the same distance from your mouth throughout.
• Read something ordinary. You do not need a script. Describe your day for thirty seconds. Natural rhythm clones better than a performance.
• Include range. A question, a statement and a short list give the model more than a flat paragraph does.
• Do not process it first. Skip the noise-removal app. Heavy denoising flattens the detail the clone is built from.

What it costs

Creator Reel quotes the whole reel before you commit — the cost of the draft clips plus a build fee — and the writing step is free. Two optional extras are priced separately and shown on their own buttons.

Step	Cost
Script writing	Free
Draft clips	Quoted from the model and length you pick, shown before you confirm
Storyboard preview	20 credits
Final build	10 credits
Thumbnail	15 credits

The clip price depends heavily on the model. Seedance 2 Fast at 720p runs at roughly 10 credits per second of finished video, while Seedance 2 at 720p is roughly 16.5 credits per second, and a longer 2-minute reel costs proportionally more than a 30-second one. Because the figure is quoted per job, read it in the tool rather than working it out in advance. Credits come from a monthly plan or from one-time packs that never expire, and failed generations are refunded. The plan tiers and per-credit rates are listed on the pricing page, and the wider pattern of how credits are spent is broken down in VdoBloom credits explained.

Whose voice you may clone

This is the part that is not a technical question. The tool’s own note says it plainly: only use your own voice, or one you have rights to. A cloned voice is a biometric identifier, and putting words into someone else’s mouth with it is a different act from imitating them badly. The safe rule is simple — if you would not be comfortable sending them the finished reel and asking for permission afterwards, do not clone it. That covers employees reading your scripts, voice actors, narrators, and anyone you found in a video online. If you need a specific tone rather than a specific person, the platform has tools designed for exactly that, and they avoid the consent problem entirely.

Tools that shape a voice without cloning anyone

Two audio tools do something related but different, and it is worth not confusing them. Voice Design creates a voice from a description — age, accent, tone, pace — which you can then use to read a script; nothing about it is derived from a real person. Voice Changer is speech-to-speech: you record or upload your own speech and hear it in another of the available stock voices, with the same words and timing. Neither is described as cloning a specific person, and neither should be treated as one. If you need a voiceover rather than a person’s voice, the text-to-speech tool is the right stop, and the mechanics of pairing a voice with a face are covered in AI voice and lip-sync on VdoBloom.

When the clone does not take

Voice conversion inside a reel is not guaranteed on every clip. Short lines, heavy background noise in the source sample, or a sample recorded on a laptop microphone several feet away all raise the chance that a clip keeps a stock voice. The tool reports this rather than hiding it. The practical fix is a cleaner thirty-second sample, and then a re-run — the failure is reported per clip, so you can tell whether the problem was the sample or one particular line.

Frequently asked questions

Is there a standalone voice cloning tool on VdoBloom?

No. Voice cloning exists only inside Creator Reel, behind the Use my voice control. It is not exposed in the text-to-speech tool, Voice Design or Voice Changer, and it cannot be used to dub an arbitrary video.

Why is the Use my voice button locked?

Either your account is on a free tier, in which case the control shows a prompt to upgrade, or cloning is currently switched off platform-wide, in which case it reads as unavailable and reels fall back to the default AI voice.

How long does the voice sample need to be?

About thirty seconds is what the recorder asks for. The upload accepts a file up to 25 MB in mp3, wav, m4a, ogg, webm, flac or aac. One clean continuous take beats a longer file full of noise.

Will the lip-sync still match my voice?

Yes when the conversion succeeds — the reel is built so the voiceover is converted to your voice without breaking the sync. If a clip cannot be converted, the reel tells you which one and keeps the default voice for it.

Can I clone a colleague’s or a voice actor’s voice?

Only with their permission. The tool’s own note limits it to your own voice or one you have rights to. Cloning a real person’s voice without consent is both a platform-rules problem and, in a growing number of places, a legal one.

Open Creator Reel, record your thirty seconds, and build a one-minute reel to hear it back before committing to anything longer. New accounts start with 10 free credits and no card, and the exact cost of a reel is shown before it renders.