I'm excited to announce SexyVoice.ai, a passion project that makes it possible to create cinematic, realistic AI voices in seconds and even clone your own voice with just a few snapshots of audio. Built with Next.js 16, Supabase, and a suite of modern tools, SexyVoice.ai is my answer to a modern, credit-based audio studio designed to be global by default.
https://github.com/gianpaj/sexyvoice
Why SexyVoice.ai exists
Voice tech feels stuck in demos: you either wrestle with glue code to stitch together cloud APIs, or you're limited to one model and one language. SexyVoice.ai brings a polished UI, a responsive dashboard, and a validated voice pipeline so you can focus on creating rather than configuring.
Highlights
- AI voice generation in 20+ languages with curated voice selections and emotion-aware outputs.
- Voice cloning with as little as 10 seconds of speech, powered by Fal.ai and Google Generative AI TTS models.
- Credit-based usage and Stripe-powered billing so you can experiment without surprises.
- Supabase authentication + profile dashboard, plus Cloudflare R2 storage for your audio history.
- International-ready UI, rate limiting for stability, and Sentry/PostHog telemetry for production confidence.
Try it and help shape it
SexyVoice.ai is live at https://sexyvoice.ai — hop in, generate a voice,
and drop feedback on the roadmap (https://sexyvoice.featurebase.app) so I can keep improving the experience.
I'll also be publishing behind-the-scenes updates on X https://x.com/gianpaj as the platform grows.
Let me know if you want a separate short post or social preview copy!
Top comments (0)