OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Audio & Voice AI

FCITX5 Vinput

FCITX5 Vinput is a Linux desktop voice-input application for Fcitx5 that transcribes speech locally or through cloud ASR and can apply LLM-based rewriting to recognized text.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

FCITX5 Vinput — OpenViST project profile in Audio & Voice AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want voice input within Fcitx5, especially if you need a choice of local or cloud recognition or want to edit dictated text through LLM scenes.

Who it is for

  • Fcitx5 users who want to dictate text into desktop workflows.
  • Linux users who want to choose between local sherpa-onnx recognition and cloud ASR.
  • Users who want spoken instructions to modify selected text or LLM scenes for correction, formatting or translation.

Documented uses

  • Dictate text using tap-to-toggle recording or hold-to-talk.
  • Switch between local and cloud speech recognition.
  • Apply a spoken instruction to selected text.
  • Post-process recognized speech for error correction, formatting or translation.
Before you choose

Requirements & limits

The repository is public and identifies the project as GPL-3.0 licensed.

  • The full version includes the local sherpa-onnx runtime; the Lite variant is cloud-only and omits that runtime.
  • Cloud ASR providers named in the README include Doubao, Aliyun Bailian, ElevenLabs and OpenAI-compatible providers.
  • GUI setup and CLI control are both provided, with the CLI described as offering full control.
  • The documented source-build dependencies include CMake, Fcitx5, PipeWire, libcurl, nlohmann-json, CLI11 and Qt6.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a local voice-input model

Start the Vinput daemon, activate a model in the GUI and use the default recording key.

Before you start

  • A Linux desktop using Fcitx5
  • The full Vinput version with local sherpa-onnx support

Steps

  1. Enable and start the user service with systemctl --user enable --now vinput-daemon.service.

  2. Restart Fcitx5 with fcitx5 -r.

  3. Open Vinput GUI, go to Resources → Models, then download and activate a model.

  4. Tap Alt_R to toggle recording, or hold it for push-to-talk.

Expected outcome

A model is activated for local speech recognition, and the default key starts or controls recording.

Guide limitations

  • The guide describes setup and recording controls; it does not specify model-specific performance or language coverage.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
xifan2333/fcitx5-vinput · Official homepage title · retained in its source language
How we review sourcesSuggest a correction