Parameters for speed vs quality

#2
by drmeir - opened

I am only interested in sentence boundaries, not the other punctuation. I can also sacrifice some accuracy, but speed is of utmost importance. Are there parameters I can set to get faster than default performance?

This model does only all 3 analytics.

I also have a model does only SBD and does so very quickly: https://huggingface.co/1-800-BAD-CODE/sentence_boundary_detection_multilang

I am working with auto-generated YouTube transcripts, which lack periods at the end of sentences. The model you cited assumes punctuated text...

Sign up or log in to comment