Open-source alternatives to Wispr Flow

7 genuinely open-source alternatives. The projects are compared by license, workflow, self-hosting, and pricing.

Why look for a Wispr Flow alternative?

Wispr Flow turns natural speech into polished text across applications. People look for alternatives when they want fully local speech-to-text, open-source software, no recurring subscription, offline dictation, custom hotkeys and vocabulary, or more control over whether audio and AI cleanup leave their device.

VoiceInk

Privacy-focused macOS dictation with local transcription.

VoiceInk is a GPL-3.0 macOS dictation application with local speech models, global push-to-talk shortcuts, custom vocabulary, application-aware formatting, and optional AI enhancement.

GPL-3.0

Pricing: Source available under GPL-3.0; official signed builds may include commercial convenience features.

Local speech recognition
Global shortcuts and custom dictionary
Application-aware formatting and AI enhancement
macOS only
Primarily optimized for Apple Silicon
Official prebuilt distribution may differ from source builds

View on GitHub

Handy

Minimal offline speech-to-text for macOS, Windows, and Linux.

Handy is an MIT-licensed cross-platform speech-to-text application built around a simple press-to-talk or toggle workflow, using local Whisper or Parakeet models.

MIT

Pricing: Free and open source under MIT; local model hardware requirements apply.

Completely local and offline
MIT licensed
Works on macOS, Windows, and Linux
Intentionally minimal
Less AI rewriting than Wispr Flow
Few context-aware and synchronization features

View on GitHub

FluidVoice

Local macOS dictation with AI-powered text processing.

FluidVoice is a GPL-3.0 dictation app combining local speech recognition with local text cleanup, punctuation, tone adaptation, long-form dictation, and context-aware formatting.

GPL-3.0

Pricing: Free and open source under GPL-3.0; optional cloud AI providers may incur costs.

Local-first architecture
AI cleanup and context-aware formatting
Multiple languages and speech models
Focused on macOS
Local models need storage and compute
More complex than basic dictation

View on GitHub

OpenWhispr

Cross-platform local dictation with cleanup and meeting transcription.

OpenWhispr is an MIT-licensed cross-platform voice tool supporting local Whisper and NVIDIA Parakeet models, system-wide input, AI cleanup, custom vocabulary, multilingual transcription, meetings, and optional cloud processing.

MIT

Pricing: Free and open source under MIT; optional hosted models or services may cost extra.

MIT licensed and cross-platform
Whisper and Parakeet support
AI cleanup, meetings, vocabulary, and 100+ languages
Broader feature set requires more configuration
Some synchronization features may depend on paid services
More than a simple dictation tool

View on GitHub

Amical

Local-first dictation and note-taking with AI formatting.

Amical is an MIT-licensed desktop dictation and note-taking application combining local speech-to-text, context-aware formatting, custom vocabulary, multilingual support, and hands-free workflows.

MIT

Pricing: Free and open source under MIT; compatible local models run on your hardware.

MIT licensed and local-first
AI formatting and custom vocabulary
Offline note-taking workflows
Younger ecosystem
Smaller community
Some advanced workflows are still evolving

View on GitHub

VoiceTypr

Offline macOS and Windows dictation with customizable formatting.

VoiceTypr is an AGPL-3.0 voice dictation application for macOS and Windows with local recognition, multilingual support, formatting modes, text cleanup, file transcription, local history, and optional providers.

AGPL-3.0

Pricing: Source available under AGPL-3.0; optional external AI providers may cost money.

Local speech recognition
macOS and Windows
Formatting modes, file transcription, and BYOK
No Linux support
Advanced AI features may use external providers
Prebuilt distribution may differ from source

View on GitHub

Jarvis

macOS voice assistant with dictation and system control.

Jarvis is an MIT-licensed macOS voice assistant that combines dictation with AI text transformations, application launching, window switching, and voice commands.

MIT

Pricing: Free and open source under MIT; optional AI provider costs may apply.

MIT licensed
Dictation and text transformations
Voice commands and application control
macOS only
More complex than focused dictation apps
Smaller ecosystem

View on GitHub

Compare Wispr Flow alternatives

Tool	Stars	License	Self-hosted	Pricing
VoiceInk	—	GPL-3.0	No	Source available under GPL-3.0; official signed builds may include commercial convenience features.
Handy	—	MIT	No	Free and open source under MIT; local model hardware requirements apply.
FluidVoice	—	GPL-3.0	No	Free and open source under GPL-3.0; optional cloud AI providers may incur costs.
OpenWhispr	—	MIT	No	Free and open source under MIT; optional hosted models or services may cost extra.
Amical	—	MIT	No	Free and open source under MIT; compatible local models run on your hardware.
VoiceTypr	—	AGPL-3.0	No	Source available under AGPL-3.0; optional external AI providers may cost money.
Jarvis	—	MIT	No	Free and open source under MIT; optional AI provider costs may apply.

Which one should you choose?

Want the closest open-source Wispr Flow experience on macOS

VoiceInk combines system-wide dictation, local transcription, global shortcuts, custom vocabulary, and optional AI text enhancement.

VoiceInk

Want completely local cross-platform dictation

Handy runs locally on macOS, Windows, and Linux with a simple press-to-talk workflow and no account or subscription.

Handy

Want the broadest feature set

OpenWhispr combines local transcription with AI cleanup, custom vocabulary, multilingual support, meeting transcription, and optional cloud models.

OpenWhispr

Want advanced local formatting on macOS

FluidVoice adds local AI text processing, filler-word removal, punctuation, tone adaptation, and context-aware formatting.

FluidVoice

Want macOS and Windows support with formatting rules

VoiceTypr combines local transcription with custom formatting modes, text cleanup, file transcription, and optional providers.

VoiceTypr

Want a lightweight open-source project

Amical takes a local-first approach to dictation, AI formatting, custom vocabulary, and note-taking workflows.

Amical

Want voice commands in addition to dictation

Jarvis extends dictation with text transformations, application launching, window switching, and system control.

Jarvis

← Browse more software