Windows Real-time Voice Dictation IME: press a hotkey to speak, text appears at cursor. Built-in offline Chinese engine; use your own cloud API key for higher accuracy. Also features AI text correction, voice commands, and end-to-end voice chat.
Release Date:
Coming soon
Developer:
Publisher:
Tags

Sign in to add this item to your wishlist, follow it, or mark it as ignored

This game is not yet available on Steam

Coming soon

Interested?
Add to your wishlist and get notified when it becomes available.

Downloadable Content For This SoftwareBrowse all (1)

 

About This Software

Press the Hotkey, Speak, and the Text Lands at Your Cursor

ORI Voice Input is a real-time voice dictation tool for Windows. Press the global hotkey and start talking — recognized text appears word by word at your cursor in real time, in any input field, editor, or chat window. It works in games, too.

Key Features

  • Live Typing: Text appears as you speak. When the engine revises earlier output, it backspaces and rewrites automatically — or you can switch to one-shot insertion at the end of each sentence.

  • No Swallowed First Words: The microphone stays warmed up, with a ~400 ms pre-roll buffer, so the start of your sentence is never lost.

  • Noise-Robust Segmentation: Sentences are split automatically based on voice energy, and ambient noise is calibrated whenever recording starts.

  • Take It Back: Press Delete or left-click to interrupt; everything not yet inserted from this session is discarded.

  • Hold to Talk: Hold the hotkey while speaking; release to stop and insert immediately.

  • Auto Stop: Stops after 10 seconds of silence, and caps a single recording at 60 seconds (both adjustable).

  • Game Mode: Built for games with custom-drawn input fields, using system-level clipboard paste.

What's Included in the Steam Version

  • Built-in offline recognition engine and dependencies — ready out of the box, zero configuration

  • Steam Workshop: community content including feature extensions, themes, fonts, phrases, and prompts

  • Steam Cloud: config files and history sync by default

  • Steam Stats: tracks only total recording time, recognized character count, and session count — no text, no audio

  • Exclusive pixel-art theme

Built-In Offline Engine

The Steam version ships with a local recognition engine and its dependencies, ready to use right after installation: no account sign-up, no keys, fully offline, and no usage fees.

Please note: the built-in offline engine currently supports Chinese speech only.

Interface & System Integration

  • Status overlay: built-in themes (Light / Dark / Minimal / Pixel / Vinyl), with adjustable font, position, size, opacity, and auto-hide.

  • History: recognized text can be logged, and a separate window lets you search the full history.

  • Automatically pauses background media playback or mutes the system while recording, then restores it afterward.

Text Injection & Compatibility

Choose between clipboard paste and simulated per-character keystrokes. Clipboard contents are automatically protected and restored.

Game Mode always uses clipboard injection to support custom-drawn input fields. If the target app runs as administrator, you can elevate with one click or always launch as administrator.

Voice Commands

When enabled, local commands such as "Listen to me, select all" and "Listen to me, undo" don't consume any LLM calls. Custom phrases let you insert preset text with a single command.

Privacy

With the built-in offline engine, all audio and recognition results are processed locally.

AI Generated Content Disclosure

The developers describe how their game uses AI Generated Content like this:

Store-page promotional materials and some in-app sound effects were generated with generative AI, then selected, edited, and reviewed by the developer before use.

System Requirements

    Minimum:
    • Requires a 64-bit processor and operating system
    • OS: Windows 10 64-bit (version 1809 or newer) or Windows 11
    • Processor: 64-bit dual-core processor (Intel Core i3 / AMD Ryzen 3 or equivalent)
    • Memory: 4 GB RAM
    • Graphics: DirectX 11 compatible graphics (integrated graphics is sufficient; used for the UI only)
    • DirectX: Version 11
    • Storage: 3 GB available space
    • Sound Card: Windows-compatible audio device (a working microphone is required)
    • Additional Notes: A microphone is required. The built-in offline recognition engine works without an internet connection; cloud engines (Alibaba Cloud, Tencent Cloud, iFlytek, Volcengine) and AI correction need your own API keys and an internet connection.
    Recommended:
    • Requires a 64-bit processor and operating system
    • OS: Windows 11 64-bit
    • Processor: Intel Core i5 / AMD Ryzen 5 or better (quad-core)
    • Memory: 8 GB RAM
    • Graphics: NVIDIA GPU with CUDA support (GTX 1050 Ti / GTX 1060 or newer) for faster offline recognition
    • DirectX: Version 11
    • Network: Broadband Internet connection
    • Storage: 5 GB available space
    • Sound Card: Windows-compatible audio device (headset microphone recommended)
    • Additional Notes: A headset microphone is recommended to avoid speaker echo. A CUDA-capable NVIDIA GPU speeds up the built-in offline engine.