Update README.md

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 56.44
+         value: 57.004356
 ---
 
 # Wav2Vec2-Large-XLSR-53-Tamil
@@ -37,10 +37,15 @@ When using this model, make sure that your speech input is sampled at 16kHz.
 The model can be used directly (without a language model) as follows:
 
 ```python
+
+!pip install datasets 
+!pip install transformers 
+
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+
 import torch
-import torchaudio
+import librosa
 from datasets import load_dataset
-from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 
 test_dataset = load_dataset("common_voice", "ta", split="test[:2%]").
 
@@ -76,10 +81,16 @@ The model can be evaluated as follows on the {language} test data of Common Voic
 
 
 ```python
+
+!pip install datasets 
+!pip install transformers 
+!pip install jiwer
+
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+
 import torch
-import torchaudio
+import librosa
 from datasets import load_dataset, load_metric
-from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 import re
 
 test_dataset = load_dataset("common_voice", "ta", split="test")
@@ -90,7 +101,6 @@ model = Wav2Vec2ForCTC.from_pretrained("Gobee/Wav2vec2-Large-XLSR-Tamil")
 model.to("cuda")
 
 chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\”\ \’\–\(\)]'
-resampler = torchaudio.transforms.Resample(48_000, 16_000)
 
 # Preprocessing the datasets.
 # We need to read the aduio files as arrays
@@ -118,3 +128,15 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
+
+**Test Result**: 57.004356 %
+
+## Usage and Evaluation script
+
+The script used for usage and evaluation can be found [here](https://colab.research.google.com/drive/1dyDe14iOmoNoVHDJTkg-hAgLnrGdI-Dk?usp=share_link)
+
+## Training
+
+The Common Voice `train`, `validation` datasets were used for training.
+
+The script used for training can be found [here](https://colab.research.google.com/drive/1-Klkgr4f-C9SanHfVC5RhP0ELUH6TYlN?usp=sharing)