Talk instead of type. Dictate into any app, say commands like "open Steam", or turn rambling speech into a polished message, prompt, translation, or voice clip. The AI runs on your PC: no account, no API key.
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.
 

About This Software

Voice Notes lets you talk instead of type. One shortcut dictates into the app you already have open. Another opens a small card that rewrites what you said, so you can check it before it is sent. Or open the launch card and just say what you want.

Dictate

Speak, and the words type themselves into notes, chat, email, or an editor. The card stays on screen and does not take keyboard focus.

  • Your words: Dictation types what you said. It does not rewrite the meaning.

  • Styles: Sentences, casual, lowercase, as heard, or a clean pass that drops fillers.

  • Hands-free edits: Say "scratch that" to take back what was just typed, or "send it" to send a message.

Compose

Hold a longer passage on a dual-pane card. Your transcript stays put. The rewrite sits above it, and that is what gets sent.

  • Ready to send: Smart message, professional, summary, and translate into dozens of languages.

  • Prompt builder: Turn a spoken idea into a clear prompt for an assistant.

  • Other voices: Pirate, Shakespeare, noir, and more. Write your own modes too.

Voice commands

Open the launch card and say what you want: "talk", "prompt builder", "Japanese", "open Steam", "play Stardew Valley", "open my document budget", "pause the music". Every command and every mode's trigger words can be changed, and you can add your own to open websites, run programs, type snippets, or press keys. Commands are spoken in English for now.

Voice clips

Talk however you like. Voice Notes tidies it up (or translates it), reads it back in a clear voice, and saves an MP3 or WAV that is already on your clipboard, ready to paste into a chat, an email, or a video editor. A speaker button reads any rewrite aloud.

Private by design

  • Speech recognition on your PC: Whisper turns your voice into text offline. On Windows, built-in Windows dictation is available too.

  • AI on your PC: The model that rewrites and translates runs on your own computer. It comes with Voice Notes, so there is nothing extra to download. No account, no API key, no subscription, and nothing you say is sent to an AI service.

  • Your voices: Voice clips use the voices installed in your operating system.

Made to fit

  • The interface is in 31 languages.

  • Pick the screen and corner for the card. A monitor is remembered by the display itself, so it stays put when Windows renumbers screens.

  • Settings follow you between PCs with Steam Cloud.

Defaults are Ctrl+Alt+Space to dictate, Ctrl+Alt+Shift+Space to compose, and Ctrl+Alt+V for the launch card. All can be changed.

AI Generated Content Disclosure

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

Voice Notes uses AI that runs on your own PC. Speech recognition turns your voice into text, and an optional on-device language model rewrites or translates your own words when you pick a compose mode, such as tidying a message or translating it. Voice clips are read aloud with the voices installed in Windows. Nothing you say is sent to an AI service, and everything the AI produces comes from your own words and is shown to you before it is used.

System Requirements

Windows
macOS
SteamOS + Linux
    Minimum:
    • Requires a 64-bit processor and operating system
    • OS: Windows 10/11 (64-bit)
    • Processor: 64-bit processor (AVX2 recommended for the AI)
    • Memory: 8 GB RAM
    • Graphics: Integrated graphics
    • Storage: 1 GB available space
    • Sound Card: Microphone required
    • Additional Notes: 8 GB RAM recommended. About 6 GB of disk space: all speech and AI models are included, nothing downloads later.
    Recommended:
    • Requires a 64-bit processor and operating system
    • OS: Windows 10
    • Processor: Intel i5 Quad Core 2nd Gen+ (or equivalent)
    • Memory: 16 GB RAM
    • Graphics: nVidia GeForce GTX 2080 Ti
    • Storage: 5 GB available space
    • Sound Card: Recommended (any type)
    • Additional Notes: 64-bit OS required
    Minimum:
    • Requires an Apple processor
    • OS: macOS 12 or later
    • Processor: Apple Silicon (M1 or later)
    • Memory: 8 GB RAM
    • Graphics: Integrated graphics
    • Storage: 1 GB available space
    • Sound Card: Microphone required
    Recommended:
    • Requires an Apple processor
    • OS: macOS Golden Gate
    • Processor: Apple Silicon (M1 or better)
    • Memory: 16 GB RAM
    • Graphics: Apple Integrated Graphics
    • Storage: 5 GB available space
    • Sound Card: Recommended (any type)
    Minimum:
    • OS: Ubuntu 22.04 LTS, SteamOS 3.0 (or equivalent)
    • Processor: 64-bit processor
    • Memory: 8 GB RAM
    • Graphics: Integrated graphics
    • Storage: 1 GB available space
    • Sound Card: Microphone required
    • Additional Notes: 64-bit OS required. PipeWire or ALSA audio. About 6 GB of disk space: all speech and AI models are included.
    Recommended:
    • OS: Ubuntu 20.04 LTS, SteamOS 3.0, Ubuntu 22.04 LTS (or equivalent)
    • Processor: Intel i5 Quad Core 2nd Gen+ (or equivalent)
    • Memory: 16 GB RAM
    • Graphics: nVidia GeForce GTX 2080 Ti
    • Storage: 5 GB available space
    • Sound Card: Recommended (any type)
    • Additional Notes: 64-bit OS required