ClonyVoiceGenerate without counting
The voice runs on your GPU: unlimited on every plan. Elsewhere, every sentence has a price.
The voice & video studio that runs on your machine
Clone your voice.
Clone your voice.
Generate without counting.
ClonyVoice runs voice synthesis on your own GPU: unlimited speech, at no extra cost, and your voice never leaves your computer. Then paste a URL — the Studio turns it into a branded video with your cloned voiceover, in about three minutes.
Signups open soon — leave your email and be notified the day it does.
Windows · NVIDIA GPU (CUDA) required · macOS on the roadmap
Local generation
Unlimited voice
Your voice stays with you
10 languages
Consent-first cloning
French company
Why creators switch
Your voice stays with you
A voice is biometric data. Voice generation is 100% local — nothing to upload, nothing to leak.
From your website to a video
Paste a URL: AI script, your images, your logo, your voice. MP4 rendered on your machine.
4 tools, 1 subscription
Cloning + TTS + videos + dubbing in 10 languages. The equivalent stack costs $50–120/month elsewhere.
How it works
Clone your voice
A few minutes of audio are enough — recorded or imported, with the speaker's consent.
Generate
Text to speech, or URL to Studio video. Listen sentence by sentence, regenerate what you want.
Export MP4 or WAV
Rendered on your machine, with subtitles and sound design for videos.
Steps 1 and 3 are 100% local. The Studio and translation use our online AI, covered by your AI budget.
From your website to a finished video
No editing skills, no stock-footage collage. The Studio reads your site and builds a video that looks like your brand — because it is your brand.
Paste a URL
ClonyVoice analyzes the page: your logo, your images, your colors, your message.
The AI writes and directs
Script, scenes, rhythm and design are composed on our servers — this is what your AI budget pays for.
Your machine renders
The voiceover is generated locally with your cloned voice, and the MP4 is rendered on your GPU. Revise the script until it's right.
About 3 minutes from URL to MP4
Why we can offer unlimited voice — and cloud tools can't
Cloud voice tools pay for GPU time on every sentence you generate. So they sell credits, meter every character, and bill overages. That's not greed — it's their architecture.
ClonyVoice generates speech on your GPU. Once the app runs, a sentence costs us nothing — so we don't count it. Voice generation is unlimited on every plan, including Free.
The only thing we meter is the only thing that costs us: the server-side AI that writes scripts, directs videos and translates dubbing. That's what your AI budget is — a transparent meter on our costs, never on your voice.
- Hours of audio per day? Same price.
- No overage bills. Ever.
- Paid plans are never blocked — even with zero AI budget left.
Your voice never leaves your machine
A voice is biometric data. With ClonyVoice, your recordings, your voice models and every audio file you generate live on your computer — nothing is uploaded to clone or to speak. Publishing a voice to the VoiceStore is a separate, explicit choice.
What does reach our servers — in plain words
Studio and dubbing use AI on our servers: the text of your scripts, the analysis of the URL you submit, and translation requests. These requests count against your AI budget and are logged for billing and abuse prevention. Your audio is not part of them.
Working offline? Voice generation keeps running without a connection for up to 48 hours between license checks.
One video. Ten languages. Still your voice.
ClonyVoice translates your script (on your AI budget), then your cloned voice speaks each language locally, synchronized to the timing of the original audio. English, French, German, Spanish, Italian, Portuguese, Russian, Japanese, Korean, Chinese.
To be precise: we synchronize the audio to the original timing. We do not alter the image or the lips.
A store of voices — and a stage for yours
Need a voice you don't have? Browse the VoiceStore and download ready-to-use voices, included from the Creator plan.
Voice artist? Publish your voice on your terms, with consent built into the process, and earn when it is downloaded. Your voice becomes an asset — one you control.
Alles was Sie für Voice AI brauchen
Vom Klonen bis zur Erstellung — eine Plattform für alles.
Jede Stimme Sofort Klonen
Erfassen Sie das Wesen jeder Stimme ab nur 3 Sekunden Audio. Nutzen Sie 1 bis 5 Aufnahmen für höhere Klangtreue. Wählen Sie den Schnellmodus für sofortige Ergebnisse oder den Präzisionsmodus mit Transkription für Studioqualität.
- Ab 3 Sekunden — bis zu 5 Aufnahmen für beste Qualität
- Automatische Mehrsprachigkeit
- Bewahrt Ton, Akzent und stimmliche Identität
Stimmen per Text Erstellen
Beschreiben Sie die gewünschte Stimme und die KI erweckt sie zum Leben. Perfekt für einzigartige Charaktere, Markenstimmen oder fiktive Persönlichkeiten, die noch nie existiert haben.
- Natürlichsprachliche Beschreibungen
- Alter, Geschlecht, Akzent anpassen
- Unbegrenzte Variationen generieren
Expressive Studio-Stimmen
Nutzen Sie unsere kuratierte Bibliothek hochwertiger Stimmen, die jede integrierte Sprache sprechen können. Vom warmen Erzähler bis zum energischen Moderator finden Sie die passende Stimme für jedes Projekt.
- 9 Premium Studio-Stimmen inklusive
- Jede Stimme funktioniert in allen 10 Sprachen
- Professionelle Qualität
Eigene Modelle Importieren
Sie haben bereits Stimmmodelle? Importieren Sie sie direkt. ClonyVoice unterstützt gängige Formate von XTTS, Coqui und anderen Frameworks. Ihre Modelle, Ihre Kontrolle.
- XTTS & Coqui kompatibel
- Unterstützung für .pth, .onnx Formate
- Einfacher Drag & Drop Import
Multi-Voice Dialoge & Video
Assign different voices to each sentence for realistic dialogues. Import scripts from .txt, .srt or .vtt files. Export as audio or as subtitled MP4 video.
- Verschiedene Stimme pro Satz
- Skript-Import (.txt, .srt, .vtt)
- Video export with subtitles (MP4)
Echtzeit-Generierung & Bearbeitung
Hören Sie jeden Satz in Echtzeit, während er generiert wird. Generieren Sie einzelne Sätze neu, ohne den gesamten Text zu wiederholen. Integrierter Video-Editor mit Mehrspurtimeline.
- Satz für Satz mithören während der Generierung
- Einzelne Sätze neu generieren
- Video-Editor mit Mehrspurtimeline
Aufnehmen, Hochladen oder Herunterladen
Nehmen Sie direkt über Ihr Mikrofon mit Echtzeit-VU-Meter auf. Laden Sie Audiodateien in jedem Format hoch. Oder fügen Sie eine YouTube-URL ein und extrahieren Sie die Stimme automatisch.
- Integrierte Mikrofonaufnahme mit VU-Meter
- YouTube-URL Audioextraktion
- Auto-Rauschunterdrückung und Whisper-Transkription
Stimmmodelle Exportieren
Speichern Sie erstellte Stimmen in verschlüsselten .clonyvoice-Paketen. Sicherer Import/Export zwischen Geräten. Projekte mit vollständiger Take-Historie verwalten.
- AES-verschlüsselte Stimmpakete
- Projektverwaltung mit Take-Historie
- Stimmen mit Kollegen teilen
Vollständige REST-API integriert
Integrieren Sie ClonyVoice in Ihren Workflow mit einer umfassenden lokalen API. Erzeugen Sie Sprache, verwalten Sie Stimmen und steuern Sie alles programmatisch.
- RESTful API auf localhost
- WebSocket-Echtzeitereignisse
- API-Schlüssel mit Ratenbegrenzung
What you should know before starting
- Windows 10/11 + NVIDIA GPU (CUDA) required — macOS is on the roadmap.
- Free plan: 720p export with a "Created with ClonyVoice" watermark and outro; paid plans: clean 1080p.
- Voice: unlimited, local. Studio & translation: online, covered by your monthly AI budget.
- Offline: the app keeps working for 48 hours without a connection.
The questions everyone asks
What do I need to run ClonyVoice?
A Windows PC with an NVIDIA GPU (CUDA). As a rule of thumb: if your machine runs a recent game, it runs ClonyVoice. macOS is on the roadmap.
Is voice generation really unlimited?
Yes. It runs on your hardware, so it costs us nothing — there is no meter to run out. Generate as much as you want, on every plan.
Then what is the AI budget for?
Only for what runs on our servers: Studio script writing and video direction, and dubbing translation. Each plan includes a monthly AI budget shown as a 0-100% gauge; your voice never touches it.
Is my voice sent to the cloud?
No. Recordings, voice models and generated audio stay on your machine. Our servers only receive the text and URL-analysis requests used by Studio and dubbing — and those are logged for billing. Publishing to the VoiceStore is always an explicit opt-in.
Can I use it commercially?
Yes, from the Creator plan. The Free plan is for personal use, and its videos carry a discreet "Created with ClonyVoice" watermark and outro.
Can I clone any voice?
You may clone your own voice, or a voice whose owner has given you explicit permission. Consent is a condition of use — and it's what the upcoming EU AI Act expects of everyone.
Compare like for like
We compare our Creator plan with same-tier monthly plans, no commitment.
| ClonyVoiceCreator — $12/Monat | ElevenLabsCreator — $22/Monat | FlikiStandard — $28/Monat | |
|---|---|---|---|
| Voice generation | Unlimited — runs on your GPU | 121,000 credits/month, overages billed | Credit-based (cloud rendering) |
| Voice cloning | Unlimited cloned voices | Included | Premium plan only ($88/mo) |
| Videos with YOUR images and logo (site analysis) | URL → MP4, ≈ 30 videos/month | — | Generic stock media |
| Dubbing / translation | 10 languages, audio sync | Separate Dubbing product | Not offered |
| Where your voice goes | Never leaves your machine | Sent to and processed in the cloud | Sent to and processed in the cloud |
| Commercial use | ✓ | ✓ | ✓ |
| Runs in the browser | — (Windows app, NVIDIA GPU required) | ✓ | ✓ |
| ¹ Prices and allowances as listed on elevenlabs.io/pricing and fliki.ai/pricing on 6 July 2026 (monthly plans, USD, before tax). Offers change: check the vendors' sites. ² "Unlimited": voice synthesis runs locally on your GPU; Studio video and translation usage is covered by your monthly AI budget (≈ 30 videos/month on Creator at current rates). ³ ClonyVoice requires Windows 10/11 and a CUDA-compatible NVIDIA GPU. macOS is on the roadmap. |
|||
Four subscriptions. Replaced by one.
Creators typically stack $50–120 a month of separate tools to do what ClonyVoice does from $12/month.
| Voice cloning & text-to-speech | $6–99 per month, metered by credits |
| Faceless & marketing videos | $25–88 per month, minutes capped |
| Dubbing & translation | $25–60 per month |
| Open voice marketplace | No mainstream equivalent |
| ClonyVoice: all four, from $12/month — with unlimited local voice generation. | |
Published prices of leading tools in each category, observed July 2026.
Start free. Upgrade when it pays off.
Ready for the EU AI Act — ahead of the deadline
From August 2, 2026, the AI Act (art. 50) requires synthetic audio and video to be disclosed. ClonyVoice was built for that world: consent-first cloning, content labeling built in, by a French company. Create with a tool designed for the rules — before they apply.
Your GPU is sitting idle. Put it to work.
Clone your voice today, free. Two voices, about 3 Studio videos a month, and voice generation you'll never have to count.
One email at launch. Nothing else, no sharing.
Weitere KI-Stimm-Anwendungsfälle Entdecken
Entdecken Sie, wie ClonyVoice die Stimmerstellung in verschiedenen Branchen und Anwendungen transformiert.
Hörbücher
Podcasts
E-Learning
YouTube
Unternehmensschulungen
Videospiele
TikTok
Dokumentationen
Synchronisation
Meditation
Werbung
Stimmerhaltung
Alle Anwendungsfälle Anzeigen →