Update README.md

README.md

@@ -90,6 +90,8 @@ print("Prediction:", processor.batch_decode(predicted_ids))
 print("Reference:", test_data["text"][:2])
 ```
 
+
+
 # Code For Evaluation on OpenSLR (Hindi + Marathi : https://filebin.net/snrz6bt13usv8w2e/test_large.csv)
 ```python
 import torchaudio
@@ -101,7 +103,7 @@ import re
 test = Dataset.from_csv('test.csv')
 
 
-chars_to_ignore_regex = '[\\\\,\\\\?\\\\.\\\\!\\\\-\\\\;\\\\:\\\\"\\\\“\\\\%\\\\‘\\\\”\\\\�\\\\।]'
+chars_to_ignore_regex = '[\\\\\\\\,\\\\\\\\?\\\\\\\\.\\\\\\\\!\\\\\\\\-\\\\\\\\;\\\\\\\\:\\\\\\\\"\\\\\\\\“\\\\\\\\%\\\\\\\\‘\\\\\\\\”\\\\\\\\�\\\\\\\\।]'
 
 # Preprocessing the datasets.
 # We need to read the audio files as arrays
@@ -132,6 +134,8 @@ test = test.map(evaluate, batched=True, batch_size=32)
 print("WER: {:2f}".format(100 * wer.compute(predictions=test["pred_strings"], references=test["sentence"])))
 ```
 
+
+
 #### Code for Evaluation on Common Voice Hindi (Common voice does not have Marathi yet)
 ```python
 import torchaudio
@@ -141,7 +145,7 @@ import numpy as np
 import re
 from datasets import load_dataset
 
-chars_to_ignore_regex = '[\\\\,\\\\?\\\\.\\\\!\\\\-\\\\;\\\\:\\\\"\\\\“\\\\%\\\\‘\\\\”\\\\�\\\\।]'
+chars_to_ignore_regex = '[\\\\\\\\,\\\\\\\\?\\\\\\\\.\\\\\\\\!\\\\\\\\-\\\\\\\\;\\\\\\\\:\\\\\\\\"\\\\\\\\“\\\\\\\\%\\\\\\\\‘\\\\\\\\”\\\\\\\\�\\\\\\\\।]'
 
 # Preprocessing the datasets.
 # We need to read the audio files as arrays
@@ -176,4 +180,6 @@ print("WER: {:2f}".format(100 * wer.compute(predictions=test_data["pred_strings"
 Link to eval notebook : https://colab.research.google.com/drive/1nZRTgKfxCD9cvy90wikTHkg2il3zgcqW#scrollTo=cXWFbhb0d7DT
 
 WER :  24.944955% (OpenSLR Hindi+Marathi Test set : https://filebin.net/snrz6bt13usv8w2e/test_large.csv)
+
+
 WER: 49.303944% (Common Voice Hindi Test Split)