Merge branch 'main' of https://huggingface.co/PereLluis13/Wav2Vec2-Large-XLSR-53-catalan into main

README.md

@@ -51,15 +51,15 @@ 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
+\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
+\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
+\treturn 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
+\tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
 
 predicted_ids = torch.argmax(logits, dim=-1)
 
@@ -87,55 +87,54 @@ processor = Wav2Vec2Processor.from_pretrained("PereLluis13/Wav2Vec2-Large-XLSR-5
 model = Wav2Vec2ForCTC.from_pretrained("PereLluis13/Wav2Vec2-Large-XLSR-53-catalan") 
 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
+\tbatch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
+\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
+\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
+\treturn 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)
+\tinputs = 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
+\twith torch.no_grad():
+\t\tlogits = 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
+\tpred_ids = torch.argmax(logits, dim=-1)
+\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
+\treturn batch
 
 result = test_dataset.map(evaluate, batched=True, batch_size=8)
 import jiwer
 
 # Chunk WER computation due to memory issues, taken from https://huggingface.co/pcuenq/wav2vec2-large-xlsr-53-es
 def chunked_wer(targets, predictions, chunk_size=None):                                          
-	if chunk_size is None: return jiwer.wer(targets, predictions)                                
-	start = 0                                                                                    
-	end = chunk_size                                                                             
-	H, S, D, I = 0, 0, 0, 0                                                                      
-	while start < len(targets):                                                                  
-		chunk_metrics = jiwer.compute_measures(targets[start:end], predictions[start:end])       
-		H = H + chunk_metrics["hits"]                                                            
-		S = S + chunk_metrics["substitutions"]                                                   
-		D = D + chunk_metrics["deletions"]                                                       
-		I = I + chunk_metrics["insertions"]                                                      
-		start += chunk_size                                                                      
-		end += chunk_size                                                                        
-	return float(S + D + I) / float(H + S + D)
+\tif chunk_size is None: return jiwer.wer(targets, predictions)                                
+\tstart = 0                                                                                    
+\tend = chunk_size                                                                             
+\tH, S, D, I = 0, 0, 0, 0                                                                      
+\twhile start < len(targets):                                                                  
+\t\tchunk_metrics = jiwer.compute_measures(targets[start:end], predictions[start:end])       
+\t\tH = H + chunk_metrics["hits"]                                                            
+\t\tS = S + chunk_metrics["substitutions"]                                                   
+\t\tD = D + chunk_metrics["deletions"]                                                       
+\t\tI = I + chunk_metrics["insertions"]                                                      
+\t\tstart += chunk_size                                                                      
+\t\tend += chunk_size                                                                        
+\treturn float(S + D + I) / float(H + S + D)
 
 print("WER: {:2f}".format(100 * chunked_wer(result["sentence"], result["pred_strings"], chunk_size=4000)))
 ```
 
-**Test Result**: 15.20 %  # TODO: write output of print here. IMPORTANT: Please remember to also replace {wer_result_on_test} at the top of with this value here. tags.
-
+**Test Result**: 14.48 %  
 
 ## Training