Skip to content

Fact · Dictation software

VoCoType-linux's home page says recognition runs locally with FunASR Paraformer, and that the user's voice never needs to leave the device.

Quote found word for word on the live source, last checked October 25, 2026

In the source's own words

“Recognition runs locally with FunASR Paraformer. Your voice, drafts, and surrounding text never need to leave the device.”

VocoType Linux: native voice typing for Linux and macOS, Leonard Li, VocoType Linux (LeonardNJU), 2026, Privacy card, No uploads.

Open the original
What it measures
The pipeline section lists local capture, voice activity detection, local ASR on CPU, and commit through the input method.
What it does not show
The page describes Chinese and mixed Chinese-English speech for the ASR step.

In the data

A row of Dictation App Features Compared: Offline, Privacy, AI, as VoCoType-linux, speech on device: 1 1 = yes, 0 = no.

App
VoCoType-linux
Feature
Speech processed on the device
Detail
FunASR Paraformer, local
As of
2026-10
Note
1 means the maker's page says the app does this.

Source

Used to calculate