bene-ges commited on
Commit
fa25fb6
1 Parent(s): cbd8ff2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -1
README.md CHANGED
@@ -4,4 +4,19 @@ language:
4
  - ru
5
  library_name: nemo
6
  pipeline_tag: token-classification
7
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  - ru
5
  library_name: nemo
6
  pipeline_tag: token-classification
7
+ ---
8
+
9
+ # Russian G2P token classification model
10
+
11
+ This is a non-autoregressive model for Russian grapheme-to-phoneme (G2P) conversion based on BERT architecture. It predicts phonemes in IPA format.
12
+ Initial data was built using Wiktionary json from https://kaikki.org/dictionary/Russian/index.html
13
+
14
+
15
+ ## Intended uses & limitations
16
+
17
+ The input is expected to consist of cyrillic letters separated by space. Real space should be replaced to underscore(_).
18
+ Note that the model was trained on single words and some short phrases.
19
+ Though it can accept longer phrases its accuracy may degrade on them.
20
+
21
+ ### How to use
22
+