Update README.md

README.md

@@ -28,14 +28,12 @@ model-index:
 # Wav2Vec2-Large-XLSR-53-Marathi
 Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Marathi using the [OpenSLR SLR64](http://openslr.org/64/) dataset. When using this model, make sure that your speech input is sampled at 16kHz. This data contains only female voices, although it works well for male voice too.
 ## Usage
-The model can be used directly (without a language model) as follows, given that you have a dataset with Marathi `actual_text` and `path_in_folder` columns:
+The model can be used directly without a language model as follows, given that your dataset has Marathi `actual_text` and `path_in_folder` columns:
 ```python
-import torch
-import torchaudio
+import torch, torchaudio
 from datasets import load_dataset
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 
-#test_dataset = load_dataset("common_voice", "{lang_id}", split="test[:2%]") #TODO: replace {lang_id} in your language code here. Make sure the code is one of the *ISO codes* of [this](https://huggingface.co/languages) site.
 mr_test_dataset_new = all_data['test']
 
 processor = Wav2Vec2Processor.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi") 
@@ -44,13 +42,13 @@ model = Wav2Vec2ForCTC.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi")
 resampler = torchaudio.transforms.Resample(48_000, 16_000) #first arg - input sample, second arg - output sample
 # 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_in_folder"])
-	batch["speech"] = resampler(speech_array).squeeze().numpy()
-	return batch
+\tspeech_array, sampling_rate = torchaudio.load(batch["path_in_folder"])
+\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
+\treturn batch
 mr_test_dataset_new = mr_test_dataset_new.map(speech_file_to_array_fn)
 inputs = processor(mr_test_dataset_new["speech"][:5], sampling_rate=16_000, return_tensors="pt", padding=True)
 with torch.no_grad():
-	logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
+\tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
 predicted_ids = torch.argmax(logits, dim=-1)
 print("Prediction:", processor.batch_decode(predicted_ids))
 print("Reference:", mr_test_dataset_new["actual_text"][:5])
@@ -58,12 +56,10 @@ print("Reference:", mr_test_dataset_new["actual_text"][:5])
 ## Evaluation
 Evaluated on 10% of the Marathi data on Open SLR-64.
 ```python
-import torch
-import torchaudio
+import re, torch, torchaudio
 from datasets import load_dataset, load_metric
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
-import re
-# test_dataset = load_dataset("common_voice", "{lang_id}", split="test") #TODO: replace {lang_id} in your language code here. Make sure the code is one of the *ISO codes* of [this](https://huggingface.co/languages) site.
+
 mr_test_dataset_new = all_data['test']
 wer = load_metric("wer")
 
@@ -71,26 +67,26 @@ processor = Wav2Vec2Processor.from_pretrained("sumedh/wav2vec2-large-xlsr-marath
 model = Wav2Vec2ForCTC.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi") 
 model.to("cuda")
 
-chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“]' 
-resampler = torchaudio.transforms.Resample(48_000, 16_000) #first arg - input sample, second arg - output sample
+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["actual_text"] = re.sub(chars_to_ignore_regex, '', batch["actual_text"]).lower()
-	speech_array, sampling_rate = torchaudio.load(batch["path_in_folder"])
-	batch["speech"] = resampler(speech_array).squeeze().numpy()
-	return batch
+\tbatch["actual_text"] = re.sub(chars_to_ignore_regex, '', batch["actual_text"]).lower()
+\tspeech_array, sampling_rate = torchaudio.load(batch["path_in_folder"])
+\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
+\treturn batch
 mr_test_dataset_new = mr_test_dataset_new.map(speech_file_to_array_fn)
 def evaluate(batch):
-	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
-		pred_ids = torch.argmax(logits, dim=-1)
-		batch["pred_strings"] = processor.batch_decode(pred_ids)
-	return batch
+\tinputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+\twith torch.no_grad():
+\t\tlogits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
+\t\tpred_ids = torch.argmax(logits, dim=-1)
+\t\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
+\treturn batch
 result = mr_test_dataset_new.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["actual_text"])))
 ```
 **WER on the Test Set**: 12.70 %  
 ## Training
-90% of the OpenSLR Marathi dataset was used for training.
+Train-Test ratio was 90:10.
 The colab notebook used for training can be found [here](https://colab.research.google.com/drive/1wX46fjExcgU5t3AsWhSPTipWg_aMDg2f?usp=sharing).