Whisper (speech-to-text)
------------------------

It is highly recommended to run this on a **GPU compute** partition supporting CUDA12.

GUI
^^^
A basic Gradio web UI can be accessed `via the Open OnDemand Applications <https://ondemand.otago.ac.nz/pun/sys/dashboard/batch_connect/sys/ood_whisper-webui_apptainer>`_.

Steps on doing basic speech-to-text transcriptions is further documented `here <https://rtis.cspages.otago.ac.nz/research-computing/hosted/stt.html#whisper>`_.

commandline usage
^^^^^^^^^^^^^^^^^
TODO