ReelVocab

Watching the videos you already watch in English is a practical way to get more English input. But at an intermediate level, idioms, phrasal verbs, slang, and context-specific meanings can interrupt the flow when you have to leave the video just to look something up. ReelVocab keeps those explanations next to the video. It reads the English captions of a YouTube video and identifies expressions whose meaning may be difficult to guess from their individual words. Each item shows what the expression means in that scene. What ReelVocab does: • Finds idioms, phrasal verbs, slang, and context-specific meanings in English captions. • Shows scene-specific explanations next to the video. • Lets you click an item to jump back to its context. • Highlights the current item automatically as the video plays. • Includes four reviewed sample videos that work without an API key. Try it without an API key: ReelVocab includes four preprocessed sample videos. Their generated results were reviewed by hand before release, so you can try the main experience immediately without creating a Mistral account. Process your own videos: To process a new video, you use your own Mistral API key with access to the model used by ReelVocab. ReelVocab prepares the captions first, and model processing starts only after you confirm by pressing Start. An important limitation: The explanations are machine-generated and can still be wrong, especially when the source captions themselves are incorrect. The bundled sample videos were reviewed by hand before release. Results generated with your own API key are not human-reviewed. ReelVocab is free and open source.