Whishper is a modern tool for automatic audio transcription and translation. It can run on both personal computers and servers, providing a convenient web interface for uploading and processing files.
In this article, we will discuss how to install Whishper on different operating systems and what you need to consider before you begin.
Important note about compatibility (AVX)
Some devices, such as Synology NAS and other processors without AVX support, may encounter problems when running Whishper. For such cases, the developers have prepared special troubleshooting instructions, which are described in detail on the page dedicated to AVX issues. It is worth reading it before installation to avoid crashes during startup.
What you need for installation
Installing Whishper requires a minimum set of tools and equipment:
- A modern computer or server that supports containers.
- Docker installed — a system that allows you to run isolated services and application components.
- That’s all you need to deploy Whishper and start using it in just a few minutes.
Quick installation using a script
The official Whishper repository has a special installation script that automates the setup process. It is available for Linux, macOS, and Windows.
After running it, the system will automatically download all the necessary components, configure the environment, and deploy services within Docker. When the installation is complete, the application will be accessible via a browser at a local address.
GPU support
Whishper can use a graphics card (GPU) to speed up audio processing and translation. If you plan to enable GPU support, we recommend that you read the guide dedicated to this feature in advance. It describes in detail which graphics card models are supported and how to enable hardware acceleration before installation.
Installation on Windows
A separate installation script is provided for Windows users. Just make sure that Docker is installed and running, and then follow the steps provided. Once the installation is complete, the application will also be available locally through your browser.
Manual installation
If you want to perform the installation manually, this is also possible. To do this, you need to:
- Download the necessary configuration files from the official repository.
- Configure the environment file (.env) by specifying the desired parameters.
- Create the directories used by the LibreTranslate translation service.
This is necessary so that the container can store language models and cache. macOS users note that sometimes it is not necessary to change the access rights to these folders — it is enough to simply create them.
After that, you can launch the containers and wait for the initialization to complete. The first launch may take several minutes as the system loads models for speech recognition and translation. Subsequent launches will be much faster because the models will already be stored locally.
Accessing the application and additional settings
Once Whishper is installed, you can open it via your browser at the local address. If you want to use your own domain, you can set up a reverse proxy. For those who plan to connect HTTPS and ensure a secure connection, there is a separate guide with detailed steps.
If you encounter problems
If errors or unexpected failures occur during installation, you can contact the project developers — just create an issue in the official Whishper repository on GitHub. The team actively supports the project and helps users resolve any issues that arise.
Conclusion
Installing Whishper is an intuitive and well-documented process. The project can be deployed on both a local computer and a server, with GPU support and flexible domain configuration. Whishper is an excellent choice for those who work with audio, interviews, podcasts, or multilingual content and want to automate the transcription and translation process.
