Update audio_tokenizer.py

audio_tokenizer.py

@@ -157,11 +157,12 @@ def process_dataset(
         repo_type="dataset",
         revision="main",
         max_workers=64,
+        token=os.environ.get("HF_TOKEN")
     )
 
     # Load dataset
     print("Loading dataset...")
-    ds = load_dataset(original_dataset, split="train")
+    ds = load_dataset(original_dataset, split="train", token=os.environ.get("HF_TOKEN"))
     ds_sample_rate = ds[0]["audio"]["sampling_rate"]
 
     # Load SNAC model
@@ -272,5 +273,5 @@ For multispeaker models, ensure your dataset has a "source" field.
 
     # Upload processed dataset
     print(f"Pushing dataset to: {output_dataset}")
-    ds.push_to_hub(output_dataset)
+    ds.push_to_hub(output_dataset, token=os.environ.get("HF_TOKEN"))
     print("Done!")