Update README.md

README.md

@@ -36,6 +36,7 @@ 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
+import librosa
 import torch
 import torchaudio
 from datasets import load_dataset
@@ -46,20 +47,20 @@ test_dataset = load_dataset("common_voice", "fi", split="test[:2%]")
 processor = Wav2Vec2Processor.from_pretrained("aapot/wav2vec2-large-xlsr-53-finnish")
 model = Wav2Vec2ForCTC.from_pretrained("aapot/wav2vec2-large-xlsr-53-finnish")
 
-resampler = torchaudio.transforms.Resample(48_000, 16_000)
+resampler = lambda sr, y: librosa.resample(y.squeeze(), sr, 16_000)
 
 # Preprocessing the datasets.
-# We need to read the aduio files as arrays
+# We need to read the audio 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(sampling_rate, speech_array.numpy()).squeeze()
+    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)