- First public release
- Whisper transcription with tiny/base/small/medium/large model support
- Persistent Whisper model cache — downloaded once, reused across sessions
- Detailed transcription progress: audio duration, wait time, current model
- Ollama local translation backend (no internet required)
- DeepSeek API translation backend
- Optional global translation context: generate topic/terminology/style prompt before translating
- Parallel translation with configurable worker count
- Output formats: SRT, MKV soft-sub, MP4 hard-burned subtitles
- Electron macOS shell with local Gradio UI