Batch Processing & Languages
Introducing batch transcription for multiple files, an optimized Whisper language selector, and a new local-first macOS architecture specification.
✨ Features
-
Batch Transcription for Video Editing: Added a utility for batch processing multiple audio and video files at once. The pipeline processes files in chunks and reuses a single model load across the batch, improving memory use on macOS. This makes bulk video processing and automated AI captioning faster.
-
Enhanced Language Selector for Captioning: Added a shared, lazy-loaded language dropdown backed by the 99 Whisper-supported language constants. The smaller component improves UI performance and makes language selection easier during captioning.
-
macOS Architecture Specification: Added a definitive, local-first master architecture specification for the macOS capture engine to our project documentation. This solidifies our commitment to privacy and performance, ensuring that advanced features like our auto silence remover and audio extraction are fully optimized for Apple Silicon to deliver studio-grade results entirely offline.