Update README.md

README.md

@@ -1,7 +1,7 @@
 ---
 language: or
 datasets:
-- common_voice 
+- common_voice
 metrics:
 - wer
 tags:
@@ -11,22 +11,23 @@ tags:
 - xlsr-fine-tuning-week
 license: apache-2.0
 model-index:
-- name: danurahul/wav2vec2-large-xlsr-pa-IN
+- name: odia XLSR Wav2Vec2 Large 2000
   results:
   - task: 
       name: Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: Common Voice pa-IN
+      name: Common Voice or
       type: common_voice
-      args: pa-IN
+      args: or
     metrics:
        - name: Test WER
          type: wer
-         value: 54.86
+         value: {54.6} 
 ---
-# Wav2Vec2-Large-XLSR-53-Punjabi
-Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Punjabi using the [Common Voice](https://huggingface.co/datasets/common_voice).
+
+# Wav2Vec2-Large-XLSR-53-or 
+Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on odia using the [Common Voice](https://huggingface.co/datasets/common_voice)
 When using this model, make sure that your speech input is sampled at 16kHz.
 
 ## Usage
@@ -39,25 +40,25 @@ import torchaudio
 from datasets import load_dataset
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 
-test_dataset = load_dataset("common_voice", "pa-IN", split="test[:2%]") 
+test_dataset = load_dataset("common_voice", "or", split="test[:2%]") 
 
-processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
-model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
+processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-or") 
+model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-or") 
 
 resampler = torchaudio.transforms.Resample(48_000, 16_000)
 
 # Preprocessing the datasets.
 # We need to read the aduio files as arrays
 def speech_file_to_array_fn(batch):
-  speech_array, sampling_rate = torchaudio.load(batch["path"])
-  batch["speech"] = resampler(speech_array).squeeze().numpy()
-  return batch
+	speech_array, sampling_rate = torchaudio.load(batch["path"])
+	batch["speech"] = resampler(speech_array).squeeze().numpy()
+	return batch
 
 test_dataset = test_dataset.map(speech_file_to_array_fn)
 inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
 
 with torch.no_grad():
-  logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
+	logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
 
 predicted_ids = torch.argmax(logits, dim=-1)
 
@@ -68,8 +69,7 @@ print("Reference:", test_dataset["sentence"][:2])
 
 ## Evaluation
 
-The model can be evaluated as follows on the Punjabi test data of Common Voice. 
-
+The model can be evaluated as follows on the odia test data of Common Voice.  
 
 ```python
 import torch
@@ -78,51 +78,47 @@ from datasets import load_dataset, load_metric
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 import re
 
-test_dataset = load_dataset("common_voice", "pa-IN", split="test") 
-
+test_dataset = load_dataset("common_voice", "or", split="test") 
 wer = load_metric("wer")
 
-processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
-
-model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
-
+processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-or") 
+model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-or") 
 model.to("cuda")
 
-chars_to_ignore_regex = '[\\\\\\\\\\\\\\\\,\\\\\\\\\\\\\\\\?\\\\\\\\\\\\\\\\.\\\\\\\\\\\\\\\\!\\\\\\\\\\\\\\\\-\\\\\\\\\\\\\\\\;\\\\\\\\\\\\\\\\:\\\\\\\\\\\\\\\\"\\\\\\\\\\\\\\\\“\\\\\\\\\\\\\\\\%\\\\\\\\\\\\\\\\‘\\\\\\\\\\\\\\\\”\\\\\\\\\\\\\\\\�]'  
+chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“]'  
 resampler = torchaudio.transforms.Resample(48_000, 16_000)
 
 # Preprocessing the datasets.
 # We need to read the aduio files as arrays
 def speech_file_to_array_fn(batch):
-  batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
-  speech_array, sampling_rate = torchaudio.load(batch["path"])
-  batch["speech"] = resampler(speech_array).squeeze().numpy()
-  return batch
+	batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
+	speech_array, sampling_rate = torchaudio.load(batch["path"])
+	batch["speech"] = resampler(speech_array).squeeze().numpy()
+	return batch
 
 test_dataset = test_dataset.map(speech_file_to_array_fn)
 
 # Preprocessing the datasets.
 # We need to read the aduio files as arrays
 def evaluate(batch):
-  inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+	inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
 
-  with torch.no_grad():
-    logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
+	with torch.no_grad():
+		logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
 
-  pred_ids = torch.argmax(logits, dim=-1)
-  batch["pred_strings"] = processor.batch_decode(pred_ids)
-  return batch
+	pred_ids = torch.argmax(logits, dim=-1)
+	batch["pred_strings"] = processor.batch_decode(pred_ids)
+	return batch
 
 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**: 54.86 %  
-
+**Test Result**: 54.6 %  
 
 ## Training
 
-The Common Voice `train`, `validation` was used for training as well as validation and testing  #
+The Common Voice `train`, `validation`, and test datasets were used for training as well as prediction and testing  
 
-The script used for training can be found https://github.com/rahul-art/huggingface_wav2vec2_punjabi/blob/main/Fine_Tune_XLSR_Wav2Vec2_on_Punjabi_ASR_with_%F0%9F%A4%97_Transformers.ipynb
+The script used for training can be found [https://github.com/rahul-art/wav2vec2_or]