🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone)

Jarvis is a voice powered AI assistant for Mac, according to its repository description.

github.com/akshayaggarwal99/jarvis-ai-assistant
🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone) homepage screenshot

About 🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone)

Jarvis is an open-source voice dictation and assistant app for macOS. The user holds the Fn key, speaks, and releases it to have cleaned, punctuated text inserted where the cursor is. The README describes an MIT license and a Mac build for Apple Silicon and Intel, with an iOS version distributed through TestFlight.

Speech recognition can run locally with Whisper or NVIDIA Parakeet, and AI post-processing can use a local LLM through Ollama. The app also supports cloud options that use Deepgram and Gemini, which require the user to supply their own API keys. The README states that the app has no tracking or telemetry, but cloud options send audio or text to those external services. Pricing is not confirmed in the source. Building from source requires Node.js 18 or later, macOS 10.13 or later, and Xcode command line tools.

Key features

  • Hold Fn to record speech and insert the transcribed text into the active application
  • Removes filler words such as "um" and "like" from dictation automatically
  • Optional grammar fixes, rephrasing, bullet-point formatting, and text generation
  • Simple voice actions such as opening YouTube or setting a timer
  • Customizable prompts for dictation cleanup, mail formatting, and assistant behavior
  • Local transcription with Whisper or Parakeet, or cloud transcription with Deepgram
Best for: Mac users who want open-source voice dictation and are willing to choose between local models and cloud API keys.
Good for
  • Open source under the MIT license, with source code available on GitHub
  • Can run speech recognition fully offline with local Whisper or Parakeet models
  • Prompts for dictation cleanup and assistant behavior can be customized
Watch out for
  • AI post-processing adds about 1 to 3 seconds of latency on typical M1 and M2 chips
  • Performance of local models depends heavily on the user's hardware
  • Some listed items, such as custom voice commands and AWS Bedrock support, are on the roadmap rather than confirmed as available

🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone) at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Platforms
iOS, macOS
Open source
Yes · Source code
Language
English
Last checked
Oct 6, 2026 · Online when checked
Listed
Oct 5, 2026

🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone) alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in AI Agents & Automation

Well-known listings from the same topic.

All 313

🎙️ OpenSource Voice Dictation Agent (Wispr Flow clone) FAQ

Can speech be processed without using cloud services?
The README says Whisper or Parakeet can run fully offline, and local LLMs can be run through Ollama. Cloud options with Deepgram and Gemini are separate and require API keys.
How do I build the app from source?
Clone the repository, run npm install, npm run build, and npm run dev. The README requires Node.js 18 or later, macOS 10.13 or later, and Xcode command line tools.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing