Merge pull request huggingface#155 from osamja/fix-minds-audio-inference

Classify minds-14 on audio rather than path

chapters/en/chapter4/classification_models.mdx

@@ -66,7 +66,7 @@ classifier = pipeline(
 
 Finally, we can pass a sample to the classification pipeline to make a prediction:
 ```python
-classifier(minds[0]["path"])
+classifier(minds[0]["audio"])
 ```
 **Output:**
 ```