s415366/punctuation_restoration
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Add image, fix link to images

Browse Source
This commit is contained in:
Filip Gralinski 2021-05-28 17:35:16 +02:00
parent d700e50c09
commit fed43103b9
2 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
.seeme.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 226 KiB

12
README.md
View File

@ -1,4 +1,6 @@
## Punctuation restoration from read text
# Punctuation restoration from read text
Restore punctuation marks from the output of an ASR system.
## Motivation
@ -8,7 +10,7 @@ Speech transcripts generated by Automatic Speech Recognition (ASR) systems typic
The purpose of this task is to restore punctuation in the ASR recognition of texts read out loud.
![](img/image001.png)
![](https://github.com/poleval/2021-punctuation-restoration-prepare/tree/main/img/image001.png)
## Dataset – WikiPunct
@ -192,13 +194,13 @@ The output to be evaluated is just the text with punctuation marks added.
Final results are evaluated in terms of precision, recall, and F1 scores for predicting each punctuation mark separately. Submissions are compared with respect to the weighted average of F1 scores for each punctuation sign.
##### Per-document score:
![](img/image003.png)
![](https://github.com/poleval/2021-punctuation-restoration-prepare/tree/main/img/image003.png)
##### Global score per punctuation sign _p_:
![](img/image005.png)
![](https://github.com/poleval/2021-punctuation-restoration-prepare/tree/main/img/image005.png)
Final scoring metric calculated as weighted average of global scores per
![](img/image007.png)
![](https://github.com/poleval/2021-punctuation-restoration-prepare/tree/main/img/image007.png)
We would like to invite participants to discussion about evaluation metrics, taking into account such factors as: