OpenWhispr

The free alternative toWispr Flow,Granola

GitHub stars10,427

🟢 Beginner⏱️ ~5 min to installYou save €12/mo

OpenWhispr is the open source alternative to Wispr Flow and Granola. Press a key, speak, and the text appears right where your cursor is, in any app: email, WhatsApp, a Claude prompt, a doc. It also transcribes your Zoom, Teams or FaceTime meetings, which it detects on its own, with who said what, and files everything into searchable notes. In local mode, with Whisper or Parakeet, which understand French, you get no word cap and your audio never leaves your machine. The first launch trap: of the three modes on offer, the default one is its own cloud, free only up to 2,000 words a week, like Wispr Flow, then paid. To never pay, pick local mode, or the cloud with your own API key.

GitHub stars10,427

10.4kView on GitHub ↗

Versus Wispr Flow: the match

Price
OpenWhispr
€0, forever
Wispr Flow
€12/mo to dictate without limits
Dictation volume
OpenWhispr
Unlimited locally or with your key; 2,000 words a week on their free cloud
Wispr Flow
2,000 words per week on the free plan
Your audio
OpenWhispr
Can stay on your machine (local mode)
Wispr Flow
Goes to their servers

Wispr Flow ↗

VS	OpenWhispr	Wispr Flow ↗
Price	€0, forever	€12/mo to dictate without limits
Dictation volume	Unlimited locally or with your key; 2,000 words a week on their free cloud	2,000 words per week on the free plan
Your audio	Can stay on your machine (local mode)	Goes to their servers
Meetings	Transcription included, on-device	You need Granola on top
AI model	Yours: local or your own key	Theirs, imposed

How to install it (even from zero)

This is the simplest page in the library: a regular desktop app. No Docker, no terminal, no server.

Paste this prompt into Claude (Code or Desktop)

It will fetch the latest version from the official GitHub repo, install it on your machine and have you pick local mode, the only one that is fully free with no limit:

Installe OpenWhispr sur ma machine (l'app open source de dictée vocale, repo GitHub OpenWhispr/openwhispr) : télécharge la dernière release depuis github.com/OpenWhispr/openwhispr/releases pour mon système d'exploitation et installe-la. Ensuite, aide-moi à la configurer : au premier lancement, ne garde pas le cloud OpenWhispr proposé par défaut, gratuit seulement jusqu'à 2 000 mots par semaine ; dis-moi, selon ma machine, si je dois choisir la transcription locale (Whisper ou Parakeet, qui comprennent le français) ou le cloud avec ma propre clé API, et vérifie avec moi dans les réglages que ce mode est bien actif, puis guide-moi pour choisir mon raccourci de dictée. Explique chaque étape simplement, je débute.

Review the GitHub repository and check whether it is safe to install given my machine and my infrastructure. Ask me as many questions as possible, using widgets, to clarify my request and be as precise as you can.

2
Allow microphone and accessibility
On first launch, your system asks for two permissions: the microphone (to hear you) and accessibility (to type for you). Accept both, Claude shows you where to click.
3
Press your key and speak
Open any app, place your cursor, press your shortcut, dictate. The text types itself. That is all.

What you actually do with it

Your messages and prompts by voice

Emails, WhatsApp, Claude prompts: you speak three times faster than you type. One key, your sentence, and the text is already written in the right place.

Your meetings transcribed, no cloud

It detects your Zoom, Teams or FaceTime calls, transcribes live and identifies who is speaking. Everything happens on your machine: a client meeting's audio goes nowhere.

A voice memo becomes a clean note

Drop in an audio file, a video or a YouTube link: it transcribes and files everything into organized, searchable notes with folders. Your second brain feeds on your voice.

Project health, at a glance

Before installing an open-source tool, one question matters: is the project alive? These scores are computed from public GitHub data. They are not made up.

87/100

Excellent

Very active, well-maintained project: go ahead with confidence.

Activity: 100; Is the code moving? (commits in the last 30 days)
Maintenance: 100; How fresh is the latest update? Fresher is better.
Community: 84; Stars and contributors: people around means help and fixes.
Maturity: 70; Project age. Old enough to be solid, still alive.
Momentum: 80; Current pace versus its usual pace: speeding up or dozing off?

The repo in numbers

GitHub stars: 10.4k
Forks: 1.4k
Contributors: 108
Commits: 2.5k
Project age: 1.4 yrs
Last commit: today
Size: 201.3 MB
License: MIT

Built with

JavaScript 53.4%
TypeScript 42%
Swift 2.7%
C 0.9%

Scores computed automatically from the GitHub API (last-30-days activity, update freshness, community size, project age). Refreshed daily.

Loucash’s take

8/10

Wispr Flow sits in my expenses tracker: I pay for it every month, so I know exactly what that line costs over a year. OpenWhispr does the same dictation (the key, you speak, text appears at your cursor) and adds what Wispr will never offer: a fully local mode where your audio stays home. The thing to know: locally, quality depends on your machine. On a recent Apple Silicon Mac it is excellent; on an older machine, use the cloud with your own API key, still subscription-free. And one trap to avoid: on first launch, the app defaults to its own cloud, only free up to 2,000 words a week. Switch to local mode, and the limit is gone.

It’s for you if: You write a lot (messages, emails, prompts) and you pay for a dictation tool, or you still type everything by hand.

Also worth a look in the library

Every Tuesday: the open source alternatives that just entered the library, what they replace and what they save you. One email, nothing more.

Free. One email on Tuesdays. Unsubscribe in one click. Your data

Join the Tuesday readers

OpenWhispr

Versus Wispr Flow: the match

How to install it (even from zero)

Paste this prompt into Claude (Code or Desktop)

Allow microphone and accessibility

Press your key and speak