Update README.md

README.md

@@ -6,11 +6,11 @@ metrics:
 library_name: nemo
 pipeline_tag: automatic-speech-recognition
 tags:
-- automatic-speech-recognition
 - speech
 - audio
 - pytorch
 - stt
+- automatic-speech-recognition
 ---
 
 
@@ -24,7 +24,7 @@ NumPy 1.21.6\
 PyTorch 1.21.1\
 NVIDIA NeMo 1.7.0
 
-```
+```bash
 pip3 install nemo_toolkit['all']
 ```
 
@@ -34,14 +34,14 @@ The model is accessible within the NeMo toolkit [1] and can serve as a pre-train
 
 #### How to Import
 
-```
+```python
 import nemo.collections.asr as nemo_asr
 model = nemo_asr.models.ASRModel.restore_from(restore_path="stt_kz_quartznet15x5.nemo")
 ```
 
 #### How to Train
 
-```
+```bash
 python3 train.py --train_manifest path/to/manifest.json --val_manifest path/to/manifest.json \
 --accelerator "gpu" --batch_size BATCH_SIZE --num_epochs NUM_EPOCHS \
 --model_save_path path/to/save/model.nemo
@@ -49,18 +49,18 @@ python3 train.py --train_manifest path/to/manifest.json --val_manifest path/to/m
 
 #### How to Evaluate
 
-```
+```bash
 python3 evaluate.py --model_path /path/to/model.nemo --test_manifest path/to/manifest.json --batch_size BATCH_SIZE
 ```
 
 #### How to Transcribe Audio File
 
 Sample audio to test the model:
-```
+```bash
 wget https://asr-kz-example.s3.us-west-2.amazonaws.com/sample_kz.wav
 ```
 This line is to transcribe the single audio:
-```
+```bash
 python3 transcribe.py --model_path /path/to/model.nemo --audio_file_path path/to/audio/file
 ```
 
@@ -81,7 +81,7 @@ In total, KSC2 contains around 1.2k hours of high-quality transcribed data compr
 
 ## Performance
 The model achieved:\
-Average WER: 13.53%\
+Average WER: **13.53%**\
 through the applying of **Greedy Decoding**.
 
 ## Limitations