Whispering
A speech-to-text app inside the Epicenter monorepo under AGPL-3.0 that records, transcribes with a provider you choose, cloud or local, optionally polishes the text, and pastes it at the cursor.
What it does
Whispering's own summary is "Press shortcut → speak → get text." It "records speech, transcribes it with a provider you choose, optionally polishes the transcript, and delivers the text". The provider can be a cloud service or a self-hosted endpoint in both the browser and desktop builds; on-device transcription with GGUF models is available only in the Epicenter desktop build, because that is where the native model runtime lives.
System-wide shortcuts and pasting at the active cursor work on desktop; the browser build falls back to the clipboard. The README says audio "leaves the device only when the selected transcription provider requires an upload".
What it replaces
Whispering replaces Wispr Flow and Superwhisper for someone who wants to pick the transcription provider themselves and keep their data in a store they own. The monorepo's stated principle is "Your data outlives the app that wrote it."
Where it falls short: neither README lists an installer to download; you build from source. Which speech models, speaker labelling and meeting capture are not described. The former standalone repository at braden-w/whispering is archived and redirects here, and the monorepo's README says the previous data stack was deleted rather than migrated.
How it runs
Clone the monorepo, bun install, then bun dev:whispering or bun dev:epicenter. Desktop builds need Rust for Tauri. Whispering does not own a native shell of its own: the README says "Epicenter owns the only Tauri runtime". There is no hosted browser deployment any more. The maintainers describe a hosted Epicenter sync service with billing and a self-host reference deployment; no prices are stated.
The licence is AGPL-3.0-or-later on every package and app. The root LICENSE is an index rather than a licence text, which is why GitHub's API reports no licence. An MIT tier existed until August 2026 and was dissolved; versions already published under MIT stay MIT.
Who it fits
A developer who wants to run dictation from source, choose the provider, and keep the transcripts in a local-first store. Someone who wants a downloadable installer should pick Handy, VoiceInk or OpenWhispr.
Limits the README states
"A hosted web runtime with a host-owned replica is refused, and so are third-party installed apps, for now." "Signed-in sync sends your data to a trusted server that reads it in plaintext." On-device transcription "is available only through Epicenter".
Sources
- Epicenter root README and LICENSE index: github.com/EpicenterHQ/epicenter, read 2026-09-16.
- Whispering README: apps/whispering, read 2026-09-16.
More voice dictation apps
A free, offline push-to-talk dictation app for macOS, Windows and Linux that runs Whisper or NVIDIA Parakeet on your own machine and pastes the text at your cursor, under MIT.
A dictation and meeting-notes app under MIT that runs Whisper or Parakeet locally or cloud models with your own key, detects Zoom, Teams and FaceTime calls, and labels speakers on device.
A macOS dictation app under GPL-3.0 whose README names Superwhisper and Wispr Flow as the tools it replaces, with local models, per-app modes, a personal dictionary and an AI assistant mode.
Added September 16, 2026. Found an error on this page? Write to reveneau@licheo.com and it gets fixed in the next weekly pass.
Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.