xifan2333

xifan2333 / fcitx5-vinput

Public

fcitx5 本地语音输入法

23
4
100% credibility
Found Mar 13, 2026 at 22 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
C++
AI Summary

fcitx5-vinput is an offline voice-to-text input plugin for Fcitx5 that enables speaking to type with optional AI refinement.

How It Works

1
🔍 Discover voice typing

You hear about a handy tool that lets you speak to type in any Linux app, no internet needed.

2
📥 Install easily

Download the ready package for your Linux and add it with one simple command.

3
🤖 Grab a speech brain

Pick and download a free voice model that understands your language perfectly.

4
⚙️ Switch it on

Enable it in your keyboard settings and start the quiet background helper.

5
🎤 Speak your words

Hold down a key like right Alt, talk naturally, release – watch text appear like magic.

6
🧠 Boost with smarts

Link a clever AI helper to auto-fix errors, translate, or format what you say.

✨ Type by voice forever

Now dictate notes, emails, or code hands-free – fast, accurate, and effortless.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 22 to 23 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is fcitx5-vinput?

fcitx5-vinput is a C++ plugin for Fcitx5 that adds offline voice input to Linux desktops. Hold Alt+Right to record speech, release to transcribe via local sherpa-onnx models, and commit text directly into any app. It pairs with fcitx5 rime github or fcitx5-mozc github setups, solving slow typing for non-English scripts without cloud dependencies.

Why is it gaining traction?

Unlike cloud-based voice tools, it runs fully local with PipeWire audio capture and optional LLM post-processing via OpenAI-compatible APIs like Ollama for error fixes, translations, or code comments. Command mode lets you select text, speak changes like "translate to English," and replace it instantly. A CLI handles models, scenes, and daemon control, plus Qt GUI for configs.

Who should use this?

Fcitx5 users on Arch or Ubuntu handling Chinese (Rime), Japanese (Mozc), or Vietnamese (Unikey) input who dictate docs or code. Devs on Wayland wanting hands-free edits in Vim or IDEs. Accessibility seekers needing reliable offline speech-to-text across apps.

Verdict

Grab it if you're deep in fcitx5 themes or android ports on GitHub—prebuilt debs and PKGBUILDs make setup easy. With 19 stars and 1.0% credibility, it's early but documented well; productionize after testing models like paraformer-zh.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.