Running Whisper With Web UI Locally on Windows Inside Docker

Introduction

This tutorial will guide you through the process of running Whisper with a web user interface locally on a Windows 11 machine using Docker. Whisper is an automatic speech recognition system that you can use for various transcription tasks. By following these steps, you will set up Docker Desktop and the Whisper Web UI, allowing you to transcribe audio files directly from your computer.

Step 1: Install Docker Desktop

To begin, you need to install Docker Desktop on your Windows 11 system.

1. Download Docker Desktop

   • Visit the Docker Desktop download page.
   • Choose the Windows version and start the download.

2. Install Docker Desktop

   • Run the downloaded installer and follow the on-screen instructions.
   • Ensure that you enable WSL 2 (Windows Subsystem for Linux) during installation, as it is required for Docker to function correctly.

3. Verify Installation

   • Open a command prompt or PowerShell.
   • Type docker --version to check if Docker is installed correctly. You should see the installed version number.

Step 2: Set Up the Whisper Web UI

Next, you will download and set up the Whisper Web UI provided by the user pluja.

1. Download Whisper Web UI

   • Go to the Whisper Web UI GitHub repository.
   • Click on the green "Code" button and select "Download ZIP".
   • Extract the downloaded ZIP file to a preferred location on your computer.

2. Open Command Prompt in the Extracted Folder

   • Navigate to the folder where you extracted the Whisper Web UI files.
   • Click on the address bar in File Explorer, type cmd, and press Enter. This will open a command prompt in the current directory.

3. Run Docker Compose

   • In the command prompt, type the following command to start the application using Docker Compose:

     docker-compose up
     

   • This command will build and run the Docker containers defined in the docker-compose.yml file.

4. Access the Web UI

   • Once the containers are running, open your web browser.
   • Navigate to http://localhost:5000 to access the Whisper Web UI.

Step 3: Using the Whisper Web UI for Transcription

After setting up the Whisper Web UI, you can start transcribing audio files.

1. Upload Audio File

   • On the Whisper Web UI, look for the upload feature.
   • Choose an audio file from your computer that you want to transcribe.

2. Select Transcription Options

   • Adjust any settings available for transcription, such as language or model settings, if applicable.

3. Start Transcription

   • Click the button to start the transcription process.
   • Wait for the process to complete; the time taken will depend on the file length and your system's performance.

Conclusion

You've successfully set up and run Whisper with a web UI on your Windows 11 machine using Docker. This setup allows you to transcribe audio files directly from your local environment. If you encounter issues with transcription accuracy, keep in mind that performance may vary based on the audio quality and model used. As a next step, consider experimenting with different audio files or settings to optimize your transcription results.