Prerequisites and Installation

Reading time ~2 minutes

To ensure optimal performance and compatibility, follow these steps when installing and configuring the WebSpeechtotext service on a Windows 10 or Windows 11 operating system:

1. Update Windows

Before proceeding, ensure your operating system is up to date. Go to Settings > Update & Security > Windows Update and install any pending updates. This helps prevent compatibility issues and ensures access to the latest system improvements.

2. Install Visual C++ Redistributable

It is required to install the latest Microsoft Visual C++ Redistributable from the official Microsoft website. Choose the latest version appropriate to your OS system architecture.

3. Install the Latest NVIDIA Drivers

Install the latest drivers for your NVIDIA GPU from the NVIDIA official website. Select your GPU model, download the latest driver, and follow the installation prompts.

4. Reboot Your Computer

After updating your system and installing the required components, reboot your machine. This ensures that all changes take effect properly.

5. Download the WebSpeechtotext package

Download the WebSpeechtotext application package and extract the contents of the archive to a folder of your choice.

For example: C:\WebSpeechtotext

6. Download the Speech Recognition Models

To enable transcription functionality, download the models (files with the .bin extension) required by Whisper:

Link: Download from Hugging Face

Note	Note that `ggml-large-v3.bin` is used as a default model, as specified in the configuration file below. However, you can experiment with other models if desired.

Place the model file into the same directory where you extracted the application. It is also possible to place the model archive in any other directory.

For example: C:\WebSpeechtotext\models