AI Product · Audio
Wave
An AI podcast studio — turn a topic into a scripted, multi-speaker episode with generated audio, organised into shows you can publish and browse.
Organisation
Independent Project
Role
Solo — full-stack
Timeline
2026
Stack
React 19 Vite Tailwind CSS TanStack Query Appwrite Express Google GenAI
Overview
Wave is a podcast studio where the recording step is generated rather than recorded. You create a show, add an episode, and the platform scripts and synthesises the audio — multi-speaker, as a finished WAV — then files it under your podcast for playback and sharing.
Key Features
- AI episode generation through an Express service on Google GenAI, producing multi-speaker conversational audio and encoding it to WAV server-side.
- Show and episode management — create podcasts, add episodes to them, and browse a per-user library of shows with dedicated podcast and episode pages.
- Modern React 19 frontend built with Vite, Tailwind CSS v4, Radix primitives and Motion, with TanStack Query handling server state, caching and background refresh across the app.
- Appwrite auth with OAuth callback flow for sign-in, sign-up and session handling, plus resizable panel layouts and toast feedback for longer-running generation jobs.
Impact
The barrier to a podcast is rarely the idea — it is the recording, editing and production. Wave removes all three from the loop, so going from a topic to a listenable episode is a form submission rather than a studio session.