rockbenben/subtitle-translator

What it solves

Subtitle Translator addresses the common issues of using general-purpose AI translators for subtitle files: the tendency of AI models to accidentally rewrite or corrupt timecodes and the tediousness of translating files one by one. It ensures that the timeline remains untouched by separating the dialogue from the timing data before translation and enables the batch processing of entire seasons of content.

How it works

The tool operates entirely client-side in the browser or via a CLI. It strips timecodes, cue numbers, and headers from subtitle files locally, sending only the dialogue text to the translation engine. After translation, it re-inserts the timing data into the translated text. It supports a wide array of translation backends, including traditional machine translation APIs (like DeepL and Google Translate) and dozens of LLM providers (such as OpenAI, Claude, and DeepSeek). For LLM-based translations, it can send surrounding lines as context to improve dialogue coherence and character consistency.

Who it’s for

Content creators, anime fansubbers, and anyone needing to translate large volumes of subtitles across different languages while preserving the exact timing and formatting of the original files.

Highlights

• Batch Processing: Translate hundreds of files simultaneously with a summary of success and failure.
• Timeline Protection: Structural separation ensures AI models cannot disrupt timecodes.
• Context-Aware LLM Translation: Uses surrounding lines as context for more natural, consistent dialogue.
• Broad Compatibility: Supports .srt, .ass, .vtt, and .lrc formats with automatic encoding detection.
• Privacy-First: All processing and API key storage happen locally in the browser; no content is sent to a central server.
• Multi-Language Output: Ability to translate a single source file into multiple target languages in one pass.
• CLI Support: A headless command-line interface for scripting and automation.