Handy vs voicebox
Push-to-talk offline dictation: hotkey, speak, text lands in whatever field has focus. Whisper or Parakeet fully on-device; cross-platform Rust/Tauri, built to be forked. — versus — Local-first AI voice studio: clone voices, generate speech via 7 TTS engines in 23 languages, dictate system-wide with Whisper, and give any MCP-aware agent a cloned voice. Tauri, MLX/CUDA.
Both give you system-wide local Whisper dictation via a hotkey in a Tauri app. Voicebox bundles it into a full voice studio (cloning, 7 TTS engines, MCP agent voices); Handy is dictation only — lighter, simpler, and deliberately forkable.
| Handy | voicebox | |
|---|---|---|
| Stars | 27k | 44k |
| Forks | 2.3k | 5.3k |
| Language | Rust | TypeScript |
| License | MIT | MIT |
| Last activity | today | 7 days ago |
| Topics | voice, local | voice |
| Curated connections | 2 | 3 |
Handy — the curator's take
The one-job dictation tool done right: a global hotkey, Silero VAD, local Whisper (GPU) or Parakeet (CPU), and the transcript pastes into any app — nothing leaves the machine. The 'most forkable, not best' positioning is honest and is the reason to pick it: small Rust/Tauri codebase, CLI flags for scripting a running instance, Homebrew/winget installs, a Raycast extension. NOT for meetings (no diarization, no summaries — that's Meetily) and not a voice suite (no TTS/cloning — that's Voicebox). Known rough edges: Whisper crashes on some Windows/Linux configs, and Wayland needs wtype/dotool for text injection.
voicebox — the curator's take
The whole voice I/O loop in one local app — ElevenLabs (output) plus WisprFlow (input) territory without the cloud. Zero-shot cloning, 50+ preset voices, paralinguistic tags via Chatterbox Turbo, a stories/timeline editor, and a global dictation hotkey with local-LLM cleanup. The agent hook is the standout: one `voicebox.speak` MCP tool call and Claude Code or Cursor talks in a voice you cloned, with per-agent voice binding and an always-visible on-screen pill so no agent speaks silently; a REST /speak covers non-MCP harnesses. NOT a library — it's a desktop product (Tauri): Linux is build-from-source, models are hefty downloads, and engine quality varies (only Turbo interprets [laugh]-style tags; others read them literally).