File size: 607 Bytes
c622037
 
 
 
 
 
 
 
 
2e82ad0
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
Based on Finnish pretrained T5 model version small-nl24

Train data:
Around 300k samples from from following datasets 
- [wikipedia](https://huggingface.co/datasets/wikipedia) 
- [Yle Finnish News Archive 2011-2018](http://urn.fi/urn:nbn:fi:lb-2017070501) 
- [Yle Finnish News Archive 2019-2020](http://urn.fi/urn:nbn:fi:lb-2021050401) 
- [Finnish News Agency Archive (STT)](http://urn.fi/urn:nbn:fi:lb-2018121001) 
- [The Suomi24 Sentences Corpus](http://urn.fi/urn:nbn:fi:lb-2020021803)

Tested with 1000 samples from the previous datasets Median CER 1.1% MEAN CER 4.2%
More detailed info coming later...