Cloud-quality voice-to-text,
on your device

No cloud. Lightweight. Built for macOS.
Works wherever you type.
Starting at $1/mo,
$10 for lifetime

Download for macOS
Free trial: Try your first 3,000 words free.
No card required.
~0.6s
speech to text
100%
on-device
4x
faster than typing
$1/mo
to get started

Full pipeline, measured p50 on an Apple M4 Max.
Real-world times vary by Mac and how long you speak. Full method.

On-device cleanup

Stream of consciousness,
polished

Injecting a raw transcript means signing your name to your unedited inner monologue. A real language model runs inside Evoglyph and polishes what you said before it lands, with nothing leaving the Mac.

Recording

What you said · raw transcript

um so yeah i think we should uh we should probably ship the the parakeet fix today because like the latency thing is is basically gone now you know

Cleaned by evoglyph

What's injected · cleaned output

I think we should ship the Parakeet fix today. The latency regression is basically gone now.

LFM2-2.6B running in-process, no server, no network call
~0.6s

Speech to finished, cleaned-up text on an Apple M4 Max, transcription and the AI pass together.

~95%

of short transcriptions finish in under 1.5 seconds. About 73% finish in under one second.

0

network calls in the path. The whole pipeline runs on the Neural Engine, fully offline.

Measured on an Apple M4 Max for a short one-sentence transcription. Latency scales with how long you speak, and slower Apple Silicon will be higher. The transcription stage alone runs in about 0.3 seconds. Full method.

How it works

Two key presses

The same hotkey starts and ends transcription, so there is nothing to hold and nothing to learn. Press once to begin, speak, then press again and clean text lands where your cursor was.

01

Press

Press your hotkey from any app to start. Global capture is system-wide, with no window to open.

/ fn
02

Speak

Speak naturally, or in a whisper. Parakeet transcribes on the Apple Neural Engine as you talk, fully on device.

~180wpm
03

Press again

Press the hotkey again. The cleanup stage runs and polished text is pasted into whatever app you are focused in.

/ fn

Where it works

Works anywhere you can type

Verified across every type of macOS app, from editors and browsers to chat and notes. These are the ones we test, and it works in the rest too.

Code & terminal

VS Code
Cursor
Terminal

Browsers

Chrome
Safari
Firefox

Chat & AI

Slack
Claude
ChatGPT

Notes & docs

Obsidian
Notion
Linear

Works in standard text inputs across macOS: code editors, browsers, chat and email, notes and docs, terminals, and ordinary text fields. If an app has no native field, Evoglyph falls back to universal text injection.

Secure fields are the one exception. Password and other protected inputs are intentionally skipped, so Evoglyph never pastes into them.

Why evoglyph

Our value proposition

Local & fast

Audio and transcripts never leave your Mac, so there is no cloud backend to leak to. Everything runs on the Apple Neural Engine, turning speech into clean, finished text in about 0.6 seconds.

Worth paying for

Evoglyph is a maintained product, not a side project. A few dollars funds frequent releases, real support, and the newest on-device models as they ship, all for a fraction of what cloud subscriptions cost.

Works everywhere

It adapts its injection method per app, so native fields, browsers, and terminals all just work. See the verified list above.

Lightweight & accurate

One native Swift binary near 150 MB, not an 800 MB Electron app. Vocabulary boosting gets your technical terms and product names right.

How it compares

Everything you need

A carefully crafted user experience that provides essential voice-to-text functionality. No feature bloat.

Feature
evoglyph
Wispr Flow
Superwhisper
On-device voice-to-text
✓
×
✓
On-device AI cleanup (default)
✓
cloud
optional
Monthly subscription
$1/mo*
$15/mo
$8.49/mo
Yearly subscription
$5/yr*
$144/yr
$84.99/yr
One-time purchase (lifetime)
$10*
×
$249
Cross-platform (beyond macOS)
Built for macOS
✓
✓
Native binary, not Electron
~150MB
~500MB
✓

* With applied at checkout (50% off, first 3,000 users)

Pricing

Pay monthly, yearly, or once

Use at checkout for 50% off (first 3,000 users).

Monthly

$1/ month50% off with

Cancel anytime.

  • All local features
  • On-device cleanup
  • Free updates while subscribed
Subscribe now

Annual

$5/ year50% off with

Save 58% vs monthly.

  • Everything in Monthly
  • Priority support
  • Best value for daily use
Subscribe now

Lifetime

$10once50% off with

Pay once, yours forever.

  • All features, forever
  • All future updates
  • No subscription
Buy lifetime

Privacy

Your voice stays on your Mac

Your audio and transcripts stay on the device. Here is exactly what runs locally, and the short list of things that ever touch the network.

On-device processing

The speech model and the cleanup LLM both run on the Apple Neural Engine. Your audio and transcripts never leave the Mac.

Every network call listed

Model download, license check, update check, and opt-in diagnostics. That is the whole list, and never your audio or transcripts.

Disk safety

History lives in a local database that stays on your machine, and you can erase it at any time from Settings.

First-party crash reporting

Crash reports are opt-in and first-party, and they carry no audio and no transcripts, ever.

Under the hood · models

Speech recognition
NVIDIA Parakeet TDT, on the Neural Engine
2.41% WER
Text cleanup
Liquid AI LFM2-2.6B (4-bit), in-process
local
One-time download
The models download once during onboarding, then the app runs fully offline.
2.1GB

Under the hood · pipeline

01 Capture audio on hotkey
02 Transcribe on the Neural Engine
03 Clean in-process with the LLM
04 Inject into the focused app

Transcribe, clean, and inject happen in well under a second for normal phrases.

Questions

Frequently asked

An Apple Silicon Mac (M1 or later) on macOS 14 or newer. The models download once on first launch (about 2 GB), then it runs fully offline. AI cleanup is built in, with no separate runtime to install.

View all FAQs

Start dictating in minutes.

Stream of consciousness, polished.
Starting at $1/mo, $10 for lifetime

Download for macOS See pricing
Free trial: Try your first 3,000 words free.
No card required.