Update app.py

app.py

@@ -18,17 +18,12 @@ YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
 device = "cuda:0" if torch.cuda.is_available() else "cpu"
 torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
 model_id = "Kushtrim/whisper-large-v3-turbo-shqip-115h"
-model = AutoModelForSpeechSeq2Seq.from_pretrained(
-    model_id, torch_dtype=torch_dtype, use_safetensors=True, token=True).to(device)
-processor = AutoProcessor.from_pretrained(model_id, token=True)
+model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, torch_dtype=torch_dtype, use_safetensors=True).to(device)
+processor = AutoProcessor.from_pretrained(model_id)
 pipe = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor,
-                chunk_length_s=30, torch_dtype=torch_dtype, device=device,
+                max_new_tokens=256, chunk_length_s=15, batch_size=16, torch_dtype=torch_dtype, device=device,
                 token=os.environ["HF"])
 
-# pipe = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor,
-#                 max_new_tokens=128, chunk_length_s=15, batch_size=16, torch_dtype=torch_dtype, device=device,
-#                 token=os.environ["HF"])
-
 @spaces.GPU
 def transcribe(inputs, task):
     if inputs is None: