A modern, AI-powered desktop application that generates high-quality subtitles for your videos and audio files. Supports multiple AI providers: Google Gemini and OpenAI.

תוכנה מודרנית מבוססת AI לתמלול אוטומטי של קטעי וידאו או אודיו. נא לשים לב שאיכות התמלול בעברית עולה משמעותית בשימוש במודל של Gemini Pro.

• 🎙️ AI Transcription: Multi-provider support — Google Gemini and OpenAI — with per-provider API key validation and a unified "Active Model" selector.
• 🆓 Google Free Tier Support: Enable Free Tier in Settings when using a Google AI Studio free key. Switches to sequential processing to stay within daily request limits — Flash handles ~20+ hrs of video/day, Pro ~1 hr/day. No payment required to get started.
• 🌍 90+ Languages: Auto-detects languages or lets you choose from over 90 options.
• 🔄 Versions & Regenerate: Create multiple subtitle versions with different settings and switch between them instantly.
• 🌐 Translate: Translate generated subtitles to another language via a text-to-text AI model while perfectly preserving timestamps.
• ✂️ Smart Processing: Automatically handles long files by splitting them into chunks with context-aware 20-second overlap and smart stitching at boundaries.
• 🩹 Gap Healing: Intelligently detects and re-transcribes missing subtitle segments that AI skipped.
• 🎬 Render Video: Burn styled subtitles directly into a video at any target resolution (Wide 16:9, Square 1:1, Vertical 9:16, or Original).
• 🎨 Global Subtitle Styling: Full per-project style control — font, font size, text color, outline/shadow effects, background box, and precise X/Y position. Per-screen-size defaults and a one-click Reset button.
• 📤 Multi-Format Export: SRT, WebVTT, and ASS with full style embedding.
• 📝 Recent Files History: Tracks your last 10 files for quick access with automatic subtitle caching for instant restoration.
• ⚡ Quality Control: Enforces reading speeds, line limits, and minimum durations for professional-looking subtitles.
• ⌨️ Keyboard Shortcuts: Control playback (Space, Arrows) and editing (Undo/Redo, Insert/Delete, Search) with global shortcuts.
• 🎬 Advanced Visual Editor: Two-tier timeline with a zoomed detail view, precision ruler, vertical grid, and a global minimap for precise navigation. Includes a Scissors tool (C) for splitting and drag-to-trim handles.
• 🔍 Search & Replace: Find and correct text across all subtitles with "Replace All" and keyboard navigation.
• ✏️ Rich Text Editing: WYSIWYG inline subtitle editing with bold, italic, underline, and per-word color markup preserved through the full pipeline.
• 👁️ Preview Mode: Toggle between subtitle list and inline video preview. Font size scales proportionally to the selected render resolution for an accurate representation of the final render.
• 🔒 Private & Local: Your files stay on your machine (except audio chunks sent securely to the selected AI provider).
• ♿ Accessible: Full ARIA support with keyboard navigation, focus traps on modals, screen reader announcements, and labeled controls.

• Node.js: v18 or newer
• AI Provider API Key (at least one):
  • Google AI Studio (Gemini)
  • OpenAI Platform (GPT, Whisper)

1. Clone the repository

   git clone https://github.com/stskr/sublibr.git
   cd sublibr

2. Install dependencies

   npm install

3. Run the app

   npm run dev

To create an installer for your OS (.dmg, .exe, or .AppImage):

# macOS
npm run build:electron

# Windows
npm run build && npx electron-builder --win

# Linux
npm run build && npx electron-builder --linux

1. Set API Key: Click Settings (gear icon), toggle the providers you want, paste an API key for each, and click Test to verify. Select your active model from the hero dropdown. If using a Google AI Studio free-tier key, also enable the Free Tier toggle under Gemini.
2. Import Media: Drag & drop a file, or click to browse. Supports audio (mp3, wav, flac, etc.) and video (mp4, mkv, mov, etc.).
3. Generate: Select your language (or leave as Auto-detect), choose your Screen Format (Wide, Square, Vertical), and click Generate Subtitles.
4. Style: Click Global Style to customize font, font size, colors, outline/shadow effects, background box, and subtitle position. Use Reset to restore per-format defaults.
5. Edit & Preview: Review subtitles in the timeline editor or switch to Preview mode to see an accurate render preview with correct font scaling per aspect ratio.
6. Export or Render: Download as SRT/VTT/ASS, or click Render Video to burn styled subtitles directly into your video.

• Model Selection: Choose from models across all enabled providers (e.g. Gemini 2.5 Flash, GPT-4o, Whisper-1).
• Free Tier (Gemini): Toggle in Gemini settings when using a Google AI Studio free key. Enables sequential processing (1 chunk at a time) to respect daily quotas — Flash: ~20+ hrs/day, Pro: ~1 hr/day.
• Screen Format: Wide (16:9), Square (1:1), Vertical (9:16), or Original — controls line length, font size defaults, and render resolution.
• Subtitle Style: Fully configurable per-project in the Global Style panel.
• Theme: The app defaults to a modern dark theme optimized for video editing.

• Core: Electron, React 19, TypeScript, Vite
• AI: Google Gemini, OpenAI (multi-provider)
• Media: FFmpeg (via @ffmpeg-installer for easy distribution)
• Styling: Vanilla CSS with comprehensive design tokens

This project is licensed under the MIT License - see the LICENSE file for details.