Whisper large-v2 model. Supports transcription in all whisper languages. Alignment is only in English for efficiency - the model has to load a separate alignment model for each language. Reach out if you need alignment in other languages.