Lingarr
A self-hosted subtitle translator that runs in Docker and automates media library translation using local Ollama models or cloud services such as DeepL.
Lingarr is a self-hosted subtitle translator for people who maintain a media library and want subtitles in another language. It automates translation of subtitle files using a service you choose, with support for both local AI and hosted translation providers.
The choice of backend determines where translation happens. Ollama lets you use models on your own hardware, and LibreTranslate can also run locally. If you choose DeepL or a cloud AI provider, subtitle text goes to that service for translation. Running Lingarr on your own server doesn't make those external services local.
Its provider support covers dedicated translation services as well as AI models. Alongside DeepL and LibreTranslate, it works with OpenAI, Anthropic, Gemini, Mistral, DeepSeek and xAI. Google, Bing, Yandex and Microsoft are also supported. That breadth gives you a choice of translation engines within the same application, rather than tying your media library to one provider. DeepL support excludes Brazilian Portuguese.
Lingarr runs in Docker, with images for both amd64 and arm64 systems. It supports MySQL, PostgreSQL and SQLite for its database. For developers who want subtitle translation inside another application, it exposes a REST API with a Swagger endpoint reference.
