Ideally, the transcription would happen the moment the short has been uploaded to the inleo front-end
Transcribing a 59 second video takes less than a second to complete using Whisper V3 Large through Groq's API, and costs about $0.015.
Once the video is posted, the transcript is posted along with it, as a comment. Or even better, the transcript is first scrubbed using the Hive ASR Dictionary word list, and then posted:
https://inleo.io/threads/view/mightpossibly/re-khaleelkazi-3c6ox7bjl?referral=mightpossibly