My own personal copy of daanelson/whisperx
This model costs approximately $0.0093 to run on Replicate, or 107 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.
This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 10 seconds. The predict time for this model varies significantly based on the inputs.
Cog implementation of whisperX, a library that adds batch processing on top of whisper (and also faster-whisper), leading to very fast audio transcription.
Forked from https://replicate.com/daanelson/whisperx using https://github.com/replicate/yolo