v0.2.5

## What's New in v0.2.5

### 🎙️ System-Wide Dictation
- **Global hotkey** — Press `⌘+⇧+Space` from **any app** to start dictating
- **Auto-paste** — Transcribed text is automatically pasted into the active text field
- **Live partial results** — See what you're saying in real-time via WebSocket streaming ASR

### 🎬 Batch Video Dubbing Pipeline
- **Full autonomous pipeline** — Drop videos, pick languages, hit go:
  `extract → transcribe → translate → generate → mix → export`
- **Live progress tracking** — Stage, percentage, current language, segment count
- **Batch Queue UI** — Redesigned with progress bars, cancel/delete buttons, download links
- **Multi-language** — Dub to multiple languages in one pass per video

### 🧪 Tests
- **23 backend tests** — Batch API endpoints + streaming ASR helpers
- **0.3s execution** — Lightweight fixtures, no GPU required

### 🛠️ Under the Hood
- Dual-mode ASR (Turbo MLX + WhisperX Accurate)
- Enhanced download progress (speed, ETA, bytes remaining)
- Status bar black flash fix
- Cold-start model preloading
- Full accessibility audit (ARIA labels, focus-visible)
- Compact UI layout improvements

### Verification
| Check | Result |
|-------|--------|
| Backend tests | ✅ 23/23 passed |
| TypeScript | ✅ Clean |
| Vite build | ✅ 1.5s |
| Rust | ✅ Compiled |

---

> [!NOTE]
> Desktop installers will be attached once CI builds complete.

**Full Changelog**: https://github.com/debpalash/OmniVoice-Studio/compare/v0.2.4...v0.2.5