From 80544737bc4338bd0cde305b8bccb7c5209e1bdc Mon Sep 17 00:00:00 2001
From: JoJoBarthold2
Date: Sat, 16 Sep 2023 15:23:56 +0200
Subject: added a method to create a txt for the ctc decoder

---
 swr2_asr/utils/tokenizer.py | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'swr2_asr/utils/tokenizer.py')

diff --git a/swr2_asr/utils/tokenizer.py b/swr2_asr/utils/tokenizer.py
index 1cc7b84..ee89cdb 100644
--- a/swr2_asr/utils/tokenizer.py
+++ b/swr2_asr/utils/tokenizer.py
@@ -120,3 +120,9 @@ class CharTokenizer:
                     load_tokenizer.char_map[char] = int(index)
                     load_tokenizer.index_map[int(index)] = char
         return load_tokenizer
+    
+    def create_txt(self,path:str):
+      with open(path, 'w',encoding="utf-8") as file:
+        for key,value in self.char_map():
+           file.write(f"{key}\n")
+        
\ No newline at end of file
-- 
cgit v1.2.3