Voice to Text Plugin with Google Voice Keyboard

In Android you can use Google Voice typing via your keyboard for free, with which you can voice type in Obsidian Android (usually a microphone button on the keyboard) while there is not such a feature or free plugin for Obsidian desktop. I understand that there is Microsoft Speech Recognition (enabled via Windows button + H) but it doesn’t work well or even support many other languages than English. Google Docs does it for free, but only in the Google Docs or Google Slides website/app. Is it possible to create a plugin that uses Google Docs voice typing option (and/or API?) but is usable outside the Google Docs site, like in Obsidian or anywhere else in Windows?
I also know that there is already an OpenAI Whisper Obsidian plugin that does it well, but it’s still weaker than Google in voice recognition for many languages, and it also requires an OpenAI API.

If that’s not possible to have a plugin doing so, does anyone know any way to use Google Docs` voice typing option outside its website, like in Obsidian Desktop?

Maybe a transcription tool based on Whisper could meet your needs. It requires a fairly powerful computer, but then you can do all the speech-to-text locally.

On my Mac, I use MacWhisper, and on Windows, I use CapsWriter-Offline.

You don’t need to limit yourself to the tools I’ve listed. Using whisper + your language + your computer system as keywords will give you a plethora of software options to choose from.

1 Like

Thank you. Do these tools work realtime?

It’s almost real-time, depending on your device’s performance. Generally, when you speak a sentence, the transcription result will be output within 1 to 3 seconds.