Fabula

Self-hosted audiobook server inspired by Spotify/Jellyfin.

Fabula is a personal audiobook streaming platform consisting of:

• Server — ASP.NET Core (.NET 10) application that scans your audiobook library, exposes a REST API, and serves the web UI.
• Web UI — React + TypeScript single-page app, built into the server's wwwroot (planned).
• Android — Kotlin + Jetpack Compose app using Media3/ExoPlayer (planned).

Status

• Phase 1 (backend skeleton) — done
• Phase 2 (web UI) — done
• Phase 3 (Android app) — next
• Phase 4 (metadata fetchers, auth) — later

Project layout

Fabula/
├── server/
│   ├── Fabula.Api/     # ASP.NET Core host + REST endpoints + serves web UI
│   ├── Fabula.Core/    # Domain model, services, library scanner
│   └── Fabula.Data/    # EF Core DbContext, SQLite, migrations
├── web/                # (planned) React + Vite + TypeScript SPA
├── android/            # (planned) Kotlin / Jetpack Compose app
└── docs/

Prerequisites

• .NET 10 SDK
• (later) Node.js 20+ for the web UI
• (later) Android Studio for the Android app

Running

Two PowerShell helper scripts live at the repo root:

Normal use — .\run.ps1

Builds the SPA into server/Fabula.Api/wwwroot and starts the .NET server. Everything runs on a single origin: http://localhost:5075.

Use this whenever you just want to listen to audiobooks.

Development — .\dev.ps1

Opens two PowerShell windows: the .NET backend on http://localhost:5075 and the Vite dev server on http://localhost:5173 with hot reload. Vite proxies /api, /openapi and /health to the backend, so the SPA and API still share an origin.

Open http://localhost:5173 for development. Changes in web/src/** are reflected in the browser instantly without rebuild. Changes in server/** require restarting the backend window.

Manual / Linux / macOS

If you prefer not to use the PowerShell scripts:

# Production build + run
cd web && npm install && npm run build
cd ../server && dotnet run --project Fabula.Api

# Or development with hot reload
cd server && dotnet run --project Fabula.Api   # terminal 1
cd web && npm run dev                          # terminal 2

On first start the SQLite database is created automatically at server/Fabula.Api/data/fabula.db.

Useful endpoints

• GET /health — health check
• GET /openapi/v1.json — OpenAPI spec (Swagger)
• GET /api/libraries — list library folders
• POST /api/libraries — add a library folder ({ "name": "Hörbücher", "path": "D:\\Audiobooks" })
• POST /api/libraries/{id}/scan — scan a library folder
• GET /api/books?search=&page=1&pageSize=50 — list books
• GET /api/books/{id} — book detail (with chapters and file list)
• GET /api/books/{id}/cover — cover image
• GET /api/stream/{audioFileId} — audio stream (HTTP range supported)
• GET /api/progress/{bookId} — current playback progress
• PUT /api/progress/{bookId} — update playback progress

Configuration

server/Fabula.Api/appsettings.json:

{
  "Fabula": {
    "DataDirectory": "data",
    "CoversDirectory": null
  }
}

DataDirectory holds the SQLite database and (by default) the covers folder. Override either with an absolute path.

Roadmap

1. Phase 1 — backend skeleton, library scanner, streaming, progress.
2. Phase 2 — web UI (library browse, player with chapter nav, progress sync).
3. Phase 3 — Android app (Media3/ExoPlayer, offline download).
4. Phase 4 — metadata fetchers (Audible, Google Books), multi-user JWT auth, playlists.