Upload a voice. Clone it. Generate unlimited scripts in your cloned voice.
Voice cloning is the bespoke version of text-to-speech. Instead of picking a voice from a library, you supply a reference recording of a voice you have rights to (your own, a team member, a licensed talent) and the model learns that voice's characteristics. Then you can generate unlimited new scripts in that cloned voice.
This is how brands maintain voice consistency across a series — record one reference, clone it, and generate new ads, intros, and voiceovers on-demand without booking a session every time. It's how creators who work with a signature voice scale their output. It's how studios dub content into multiple languages using a single-voice reference across all regions.
The reference needs to be clear and representative of the voice's normal range. A 30-second to 2-minute sample works well. The longer the reference, the more nuance the clone captures. Once cloned, generate new scripts in a few seconds.
A critical note on consent: voice cloning requires explicit permission. Skrrol's terms restrict cloning to voices you own or have clear consent to use. Cloning public figures, celebrities, or third parties without permission isn't legal in most jurisdictions and isn't allowed on Skrrol.
Multilingual support is powerful — clone once in English and generate the same script (or different scripts) in Spanish, French, German, Mandarin, and dozens of other languages, all in the cloned voice. Skrrol's studio handles the synchronisation so a cloned voice sounds consistent across every language variant.
Upload 30s–2m of reference audio and the model learns that voice's characteristics.
Once cloned, generate as many new scripts as you want in that voice.
Generate the same cloned voice across dozens of languages — localise without losing brand voice.
Adjust speed, stability, and similarity to refine the read even after cloning.
Maintain the same voice personality across ads, intros, promos, and long-form content.
Cloned audio lands on the timeline ready for ducking, EQ, layering with music, and export.
Result: Three ads in your CMO's voice without recording sessions — useful for campaign variants and regional adapts.
Result: Five localised versions in the same voice — launch globally with consistent brand voice.
Result: Scale narration output without recording daily — clone once, generate on-demand.
Result: Extended use of a single talented voice without multiple booking sessions — cost-effective for high-volume projects.
Sign in, click Generate, pick the Voice tab, and choose Voice Cloning.
Record or drop a 30s–2m sample of the voice you want to clone. Clear audio, representative of normal range.
Give the clone a memorable name — usually the person's name or a descriptor.
Paste a script and hit Generate. The model synthesizes in the cloned voice. Re-generate single lines without redoing the whole script.
Generate the same script (or different scripts) in other languages — the cloned voice handles all of them.
Drop the audio on the timeline, adjust timing, add music, and export.
Skrrol AI uses VL credits across all generators - image, video, voice, and music. The same credit pool applies; heavier modalities (video) use more credits per generation than lighter ones (image, voice). Choose a plan and use credits across any generator.
Trial credits to clone one voice and generate a handful of scripts.
8000 VL credits — clone multiple voices, generate hundreds of scripts per month. Suitable for brands and small creators.
17000 VL credits — clones for your whole team, multilingual generation, high-volume localization.
35000 VL credits — production-scale cloning and generation for studios, agencies, and content platforms.
Only voices you own or have explicit written permission to use. Cloning public figures or third parties without consent is prohibited — it's both a legal requirement in most jurisdictions and a Skrrol policy.
30 seconds to 2 minutes. Longer samples capture more nuance; shorter samples are faster. Clear audio with minimal background noise works best.
Very accurate for the speaker's general voice characteristics. ElevenLabs' cloning is professional-grade — used in audiobooks, podcasts, and brand work.
Yes, as long as you own or have explicit permission to use the original voice. Paid Skrrol plans cover commercial creator use; the underlying provider's terms apply.
Yes — clone once and generate new scripts in dozens of languages. The voice stays consistent across all of them.
Cloning uses credits just like standard text-to-speech, calculated by script length. Standard at €7.99 covers hundreds of cloned scripts per month.
Every generation opens directly in the Skrrol Studio. These features are especially useful as the next step after a ai voice cloning generator run.
Skrrol AI runs every generator alongside a complete pro studio. Your work stays on your device. Start free.