❌ The large model can eat 6-10 GB RAM + VRAM. Older Windows machines will struggle.
✅ From tiny (fast, less accurate) to large (slower, near-human accuracy). GUI lets you pick before transcribing. whisper gui windows
✅ TXT, SRT, VTT, TSV—ready for subtitles or documentation. ❌ The large model can eat 6-10 GB RAM + VRAM