Common-voice: [nl] Create initial dataset of 500 Dutch sentences

[pcmill]

I would love to get the Dutch language going for this project. Like @highsource I have made a fork and will be making a list of sentences.

Most of the sentences will be translated from the English list, but I will also try to make some originals ;)

[highsource]

Hey @pcmill how is it going?

I wonder, what is your method to gather/create sentences?

[pcmill]

I mostly (roughly) translate sentences from English. I made a list of categories and kind of filled them:

• General
• Foods and Drinks
• Months and Times
• Medical
• Nature
• Religion
• Social
• Sports
• Proverbs
• Technologie
• Business

My sentences got merged this day actually, but I think I can add some more sentences that reference more locations in Belgium since they also speak Dutch.

[danielsjf]

@pcmill today also another big merge was done for Dutch sentences, also in collaboration with Sjoerd. There I also included many locations, but the more the better.

[danielsjf]

Just saw that the limit has gone up from 500 sentences to 5000. Therefore we need an additional 1500 for the launch. @pcmill @sroet, interested in setting up a different collaboration channel?

[pcmill]

Sure, sounds like a good idea.

[highsource]

@danielsjf Where did you saw about the 5k limit?

[danielsjf]

If you go to the website, go to languages, click on the tab progress and you will see it. Under Dutch, it says 3600/5000.

[highsource]

@danielsjf Ok, I see, thanks. Russian is at 0 at the moment, I'm working on a fork, wasn't merged yet.

But it good to know which volume is required. Wasn't apparent for me before.

[sroet]

@danielsjf @pcmill Sorry, I will not be able to contribute constructively anytime soon. (currently in the middle of moving countries) I should be able to proofread, if required.

[pcmill]

No problem @sroet, I will try my best to come up with some more sentences. You already made most of them anyway.

[pcmill]

We are live on the website with Dutch, so I can close this issue :smile:

[danielsjf]

Finally! Thanks for the great work guys!

[Gregoor]

Awesome stuff y'all!

[danielsjf]

Submitted the news on Tweakers and now we are already at 2h30 recorded and 130 speakers ;-)

[mikehenrty]

Dutch is one of our fastest growing datasets right now! Great work everyone!