Read Frog is a browser extension that turns web reading into a language learning activity through bilingual translation, text-to-speech, and vocabulary flashcards. Users select from AI providers and translation engines, including local options like Ollama.
Project overview
It combines immersive translation, context-aware AI processing, and spaced-repetition flashcards directly within the browser.
Project type
Infrastructure
Use cases
Translation & Subtitles
Deployment
Refer to project documentation
License
GPL-3.0
Best for
General users and language learners seeking an integrated browser-based tool for web page translation and vocabulary acquisition.
Key capabilities
Turn selected text into reusable AI tools with custom prompts and structured output fields, running actions directly from the selection toolbar.
Save vocabulary, definitions, and reading notes to Notebase, turn them into flashcards with customizable templates, and review due cards using a spaced-repetition scheduler.
Switch between bilingual mode (original text alongside translation) and translation-only mode, automatically re-translating visible content on mode switch without refreshing.
Select any text on a webpage to reveal a smart toolbar offering real-time translation, detailed explanations, and text-to-speech.
Extracts the page title and a concise Markdown version of the content to provide context to AI for accurate and contextually-appropriate translations.
Translate YouTube subtitles directly in the video player with translations displayed alongside the original subtitles.
Listen to any selected text with high-quality AI voices powered by Edge TTS, featuring automatic language detection, rate/pitch/volume adjustment, and smart sentence-aware chunking.
Save up to 70% on API costs by grouping multiple translation requests into single API calls with smart retry logic and exponential backoff.
Limitations and risks
AI providers can be configured with custom endpoints and API keys. Remote inference and paid services may incur costs depending on the configured provider.
The data boundary is external_optional, as external AI providers and translation services can be configured. Local orchestration and local inference depend on the user separately configuring a local model provider.
Getting started
Install the extension directly from the Chrome, Edge, or Firefox Web Store without compilation or terminal usage.