Large Romanian Dataset Formating Help

Common Voice
#1

Hi,

So one of my friends was nice enough to pass on the Romanian subtitles that they had written for a bunch for of movies over several years. The person gave me written approval to publish the sentences in the public domain, however the format of these files are problematic. I have about 9 MB of text files but it would greatly help if you guys could help me format the files in a way so we could submit it to the common voice project. The total amount of lines is roughly 289 348 & there’s 817 files. If anyone would like to help it would be very appreciated.

Thanks

#2

Hi,

Can you upload the files to github or similar so people can check the current format and suggest options?

Thanks!

#3

@nukeador I will shortly.

1 Like
#4

Okay so it wasn’t shortly, but I dropped them here: https://github.com/missuniverse/Romanian
@nukeador

#5

@dabinat might be able to help you split sentences to be able to submit them to the sentence collector.