deepmultilingualpunctuation

deepmultilingualpunctuation copied to clipboard

Can you improve it for SRT subtitle file? the SRT subtitle format like this: 1 00:00:00,000 --> 00:00:04,000 Guess how much I love you 2 00:00:04,000 --> 00:00:11,000 One autumn...

bk111

Firstly, many thanks for your awesome project which can help me do nice punctuation, but for transformer pipeline, I hope I can pass in some config to make it works...

kaka1909

Added the option to send a model path as a parameter

1

This will help if you do not want to redownload all the model files (which is nearly 2.5 GB min) Just send the local path as a parameter @oliverguhr

OoFa99

Full stops after numbers unnoticed, extra ones predicted

1

Hi and thanks a lot for the great tool! Seems that in the original punctuation removal step, punctuation in numbers is [intentionally kept](https://github.com/oliverguhr/deepmultilingualpunctuation/blob/8138125f74a8c2f67a6bce0425249ca6d108dc11/deepmultilingualpunctuation/punctuationmodel.py#L15). Perhaps due to decimal point issues or...

alexdiment