Voice Anywhere - A floating mic that turns your speech into text anywhere
by•
Voice Anywhere is an AI speech-to-text app that works everywhere. Apps, websites, coding IDEs. If you can type there, you can dictate there. A floating, pinnable mic stays above all windows so you never lose it. Fast on-device recognition, 100+ languages, and optional AI engine. Made for founders and vibe coders who move fast.
Pro tip: Use "SHIFT + R" to toggle on/off.
Replies
iSwift.dev
@nardibraho Congratulations on the launch! What level of control do users have over transcription formatting?
iSwift.dev
Why use this when Apple has dictation available?
- This app supports multiple languages, and uses AI as a fallback whenever Apple's speech recognition engine is not available (due to older devices or other reasons).
iSwift.dev
@t0ny_ns It works as a toggle, you turn it on and it remains on. Thanks for the feedback, the press to speak experience like the one you described will be something I'll add.
And yes, there's a shortcut to turn on/off dictation (SHIFT+R). Again, it stays on until you turn it off.
@t0ny_ns @nardibraho Mid-spec or mid-debug, I don't want to hunt for a mic window or bounce to Notes just to dump a thought. Voice Anywhere staying on top with a shortcut like Shift+R hits that pain point, as long as the mic can live on a screen edge and not block controls. Push-to-talk hold-to-speak and a code mode for new line and indent would cut the cleanup in IDEs. Since you default to on-device, a clear badge when AI fallback kicks in would keep trust high.
Does this only work on Apple devices? Also, is there a plan for this to be optional for mobile users or will this strictly be for desktops?
iSwift.dev
@cbutler96 For now yes, but coming to PC's and mobile soon. Even users without a mic on their PC will be able to use this by connecting their phone.
Congrats on the launch — love the “type anywhere” approach to voice dictation on macOS, it feels like a superpower for writers and builders.
The floating mic that stays on top is such a small detail but makes a huge difference. Biggest friction with dictation tools is losing the mic window behind other apps and having to hunt for it.
I do a lot of AI prompting and describing tasks verbally tends to produce better results than typing for some reason - probably because I explain more naturally. Having this work directly in VS Code or Cursor would be great for that workflow.
Does the AI fallback kick in automatically, or do you have to manually switch when Apple's engine struggles?
iSwift.dev
@philip_sorensen Thank you and glad you noticed that! It does make a difference.
For popular languages, AI can be turned on by toggling the mode from "lighting" to "brain/thinking". The latter is AI mode (uses Google's latest speech recognition model).
For some languages AI is the only model available, but the lighting/brain mode simply change how fast the speech is processed (faster = worse speech rec).
Wow, Voice Anywhere looks incredible! The floating mic that stays on top of everything is genius. Curious, does the on-device recognition work offline as well?
iSwift.dev
@jaydev13 Thank you! And yes it should work offline too
I've been using Wispr Flow with the hold-to-talk approach, and the toggle mode here is interesting—different mental model. The floating mic is smart for context switching, but I'm wondering about accuracy in noisy environments. Does the on-device Apple recognition handle background noise better than the AI fallback, or is it purely a speed/privacy tradeoff? Also, does SHIFT+R work globally across secure inputs like password fields?