Compare

Presspeech vs VoiceInk

These are philosophical neighbors: both open source, both local-first, and both can run NVIDIA's Parakeet models through FluidAudio. VoiceInk is a configurable workspace with many models and AI enhancement. Presspeech is one fixed pipeline tuned for latency.

Question	Presspeech	VoiceInk
Primary workflow	Hold hotkey, speak, release — the verbatim transcript pastes at the cursor about 100 ms later.	Hotkey dictation with optional AI enhancement, per-app Power Mode profiles, and audio/video file transcription.
Engine path	Local Parakeet TDT v3 CoreML model on the Apple Neural Engine via FluidAudio.	Your choice: Parakeet V2/V3 via FluidAudio, Whisper models via whisper.cpp, Apple Speech on macOS 26+, or optional cloud providers.
Privacy posture	No cloud path exists: no account, no telemetry, transcript history in memory only.	Local by default — “your voice data never leaves your Mac” — with optional cloud transcription and cloud enhancement if you enable them.
Price and license	Free and MIT licensed.	$25–$49 one-time for the signed, auto-updating binary; GPL-3.0 source (per its README) you can build yourself for free.
Footprint	2.5 MB app, ~80 MB RAM idle, 0% CPU between dictations; published benchmark.	13.7 MB download; resource figures not published, 8 GB RAM recommended.
Configurability	Deliberately minimal: hotkey, trigger style, language pinning, local text corrections.	Extensive: model management, AI modes with custom prompts, per-app profiles, custom vocabulary, transcription history with audio replay.

Choose VoiceInk if you want one local tool that also handles AI cleanup, per-app behavior, and file transcription, and a one-time price fits. Choose Presspeech if you want the smallest possible utility that does exactly one job — both apps can run the same Parakeet model, so this choice is about surface area, not engine.

Sources: Presspeech source, Presspeech benchmarks, VoiceInk official site and repository (checked June 2026).