Update README.md

README.md

@@ -40,7 +40,7 @@ import torchaudio
 from datasets import load_dataset
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 
-test_dataset = load_dataset("csv", data_files={"test": "/content/test.csv"}, delimiter="\t")["test"] # Adapt the path to test.csv
+test_dataset = load_dataset("csv", data_files={"test": "/content/test.csv"}, delimiter="\\t")["test"] # Adapt the path to test.csv
 
 processor = Wav2Vec2Processor.from_pretrained("csikasote/wav2vec2-large-xlsr-bemba") 
 model = Wav2Vec2ForCTC.from_pretrained("csikasote/wav2vec2-large-xlsr-bemba") 
@@ -59,7 +59,7 @@ 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.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
 
 predicted_ids = torch.argmax(logits, dim=-1)
 
@@ -86,7 +86,7 @@ processor = Wav2Vec2Processor.from_pretrained("csikasote/wav2vec2-large-xlsr-bem
 model = Wav2Vec2ForCTC.from_pretrained("csikasote/wav2vec2-large-xlsr-bemba") 
 model.to("cuda")
 
-chars_to_ignore_regex = '[\\,\\?\\.\\!\\;\\:\\"\\“]'  
+chars_to_ignore_regex = '[\\\\,\\\\?\\\\.\\\\!\\\\;\\\\:\\\\"\\\\“]'  
 #resampler = torchaudio.transforms.Resample(48_000, 16_000)
 
 # Preprocessing the datasets.