VideoCaptioner
Transcribes speech, optimizes text, translates languages, and burns styled subtitles directly into video files.
VideoCaptioner is an AI-powered video subtitling tool for Claude Code that automates the entire subtitle lifecycle from speech transcription to hardcoded video rendering. It leverages speech-to-text engines like Whisper and Bijian, supports free and LLM-driven translation across dozens of languages, fixes timing and punctuation with AI re-segmentation, and renders stylized subtitles (including anime outline and modern rounded background modes) directly into videos using FFmpeg.
Key Features
01Transcribe video and audio to SRT or JSON using free engines or Whisper API
02Translate subtitles into over 30 languages using Google, Bing, or LLM reflective translation
03Optimize subtitle text, fix ASR errors, and re-segment sentences with LLMs
04Burn styled subtitles directly into videos with custom fonts, colors, and rounded boxes
05Execute full end-to-end processing or granular step-by-step pipeline workflows
0615,879 GitHub stars
Use Cases
01Generating bilingual or translated hardcoded subtitles for international video releases
02Extracting and transcribing audio tracks from podcasts or videos into accurate text formats
03Fixing automated transcription errors and styling video captions with custom branding