Voice2Captions is a CEP extension for After Effects that automatically transcribes and subtitles audio using AI. It converts speech into multilingual captions across 99+ languages, working directly inside your timeline. The extension supports Whisper-based transcription and, from version 2.1.0, a Deepgram Nova-3 engine, with optional translation of transcribed content into up to 99 languages.
Captions can be built as single-layer transcription with timeline markers or as multi-layer subtitles for styling and animation. Version 2.1.0 added editable transcription before captions are created, smart line splitting from 1–10 words across 1–2 lines, and built-in text animation presets for After Effects. Finished subtitles export as .srt files, and FFmpeg handles internal audio file conversion.
Transcription runs through either Groq API keys, which are free, or OpenAI API keys, which are paid. The extension works on Windows and macOS, including Intel and Apple Silicon M1, M2, and M3 chipsets, and version 2.1.0 also added Premiere Pro support.