Be My Anchor 🎙️
Transform articles, tech blogs, and research papers into an engaging, hearing-first multimedia podcast experience.
Features (Milestone 1)
- High-Fidelity Neural Audio Generation: Multi-provider TTS architecture powered by Microsoft Edge Neural Voices (free, zero-key) with OpenAI TTS interface ready.
- Smart Text Chunking & Audio Stitching: Automatically splits long articles and paragraphs into natural spoken chunks, stitches them into a unified
.mp3, and calculates precise timing cues. - Local File System Audio Storage: Decoupled audio management with native HTTP Range streaming support for smooth scrubbing.
- Synchronized Web Player:
- Interactive player with Play / Pause, Progress scrubbing, Speed controls (0.8x - 2.0x), and ±15s jumps.
- Live synchronized transcript that auto-scrolls and highlights the active spoken paragraph.
- Click-to-seek directly from transcript.
- Episode Library: Local database persistence via Prisma & SQLite (easily migratable to MongoDB / PostgreSQL).
Tech Stack
- Framework: Next.js 16 (App Router) + React 19 + TypeScript
- Styling: Tailwind CSS + Lucide Icons
- Database & ORM: Prisma + SQLite
- Audio Engine:
msedge-tts(Neural TTS)
Quick Start
1. Install Dependencies
npm install
2. Initialize Database
npx prisma db push
3. Run Development Server
npm run dev
Open http://localhost:3000 in your browser.
Architecture & Roadmap
Refer to Product Requirements Document (docs/prd.md) for full architecture details:
- Phase 1 (Complete): Audio generation core, chunking, stitching, local audio storage, and synchronized web player.
- Phase 2 (Upcoming): Hearing-front LLM transformation (Feynman Deep-Dive Presenter & Dual-Host Banter).
- Phase 3 (Upcoming): Synchronized visual stage (displaying diagrams, figures, and code in sync with audio).