Clone voice,pay only for what you use: Clone any voice in minutes — pay only for what you use

I'm a backend developer, and I built Resona because voice tooling felt stuck between two bad options: subscriptions that charge you monthly whether you use them or not, or "free" tiers that hide the real limits until you hit them. Resona is a voice console: clone a voice from a short sample, generate speech in 83 languages, transcribe audio, and design new voices from a text description. It runs on a third-party voice engine behind a Cloudflare relay — you get top-tier engine quality without maintaining a single GPU. The pricing is deliberately simple: no free tier, no subscription. Credit packs start at $9, credits never expire, and every feature is metered transparently — 250 credits per clone, 1,500 per 1M UTF-8 bytes of speech, 0.15 per second of transcription, 1 per voice design. Every deduction is itemized in your dashboard, so you always know exactly what you paid for. What started as a quick relay script became a full product when I realized the console itself — browsing 200,000+ ready-made voices and hearing samples before spending credits — was the part people actually needed. I'd love your feedback: what are you building, and what's missing? Happy to answer everything below.