Update README.md

README.md

@@ -21,7 +21,7 @@ Creator & Uploader: Jooyoung Lee ([email protected])
 
 This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on a phonetically balanced native Korean read-speech corpus.
 
-## Training and Evaluation Data
+# Training and Evaluation Data
 
 Training Data
 - Data Name: Phonetically Balanced Native Korean Read-speech Corpus
@@ -33,7 +33,7 @@ Evaluation Data
 - Num. of Samples: 6,000
 - Audio Length: 12 Hours
 
-## Training Hyperparameters
+# Training Hyperparameters
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
@@ -48,22 +48,22 @@ The following hyperparameters were used during training:
 - num_epochs: 20 (EarlyStopping: patience: 5 epochs max)
 - mixed_precision_training: Native AMP
 
-## Evaluation Result
+# Evaluation Result
 
 Phone Error Rate 3.88%
 
-## Output Examples
+# Output Examples
 ![output_examples](./output_examples.png)
 
-## MFA-IPA Phoneset Tables
+# MFA-IPA Phoneset Tables
 
-# Vowels
+## Vowels
 ![mfa_ipa_chart_vowels](./mfa_ipa_chart_vowels.png)
 
-# Consonants
+## Consonants
 ![mfa_ipa_chart_consonants](./mfa_ipa_chart_consonants.png)
 
-## Framework versions
+# Framework versions
 
 - Transformers 4.21.3
 - Pytorch 1.12.1