“Gboard-like” text insertion capability?

[geraldblank]

I used to play around with Apple II BASIC and Atari BASIC, around 1980.
Speaking as a caveman stepping into the 2020s, the scope of DreamServer blows me away! I’m grateful for all the effort that went into this. Thank you!

I can’t wait to try it on my new Strix Halo/96 GB RAM Windows box.

Due to my stiff typing fingers, I’m looking for an alternative to Windows Voice (which will be useless crapware until MS figures out how to monetize it).

I was previously interested in “Handy”, until I realized its developers aren’t pursuing Strix Halo optimization: (https://github.com/cjpais/Handy/blob/main/README.md)

My short, sweet feature request:

A way to insert Whisper voice-generated text (and keyboard shortcuts) into Windows apps (instead of just the browser interface).

Put another way: Functioning like Gboard for Android phones, where one can tap the mic icon and speak into any app’s text fields.

Another feature Gboard has (which Handy lacks) is its plain-text custom user dictionary, which is a huge time-saver (except when Gboard ignores it – LOL).

If these capabilities can already be integrated into DreamServer, please advise.

Thanks again!

[rishibaghel25]

Thanks for the suggestion!

For system-wide voice dictation, I'd recommend using a dedicated desktop application instead of DreamServer itself.

DreamServer's role is to run and serve AI models (e.g., Whisper for speech-to-text). Injecting text into arbitrary Windows applications is a separate concern that is typically handled by OS-level utilities, keyboard/IME software, accessibility tools, or dedicated dictation applications.

In other words, DreamServer can provide the transcription, but getting that text into any focused application is best handled by a client application designed specifically for desktop integration.

Keeping those responsibilities separate also helps keep DreamServer focused on model serving rather than platform-specific UI and input automation features.