Update app.py

app.py

@@ -1,9 +1,39 @@
 import gradio as gr
-import whisper
+import torch
+from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
+
+
+
+
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+
+model_id = "openai/whisper-large-v3"
+
+model = AutoModelForSpeechSeq2Seq.from_pretrained(
+    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
+)
+model.to(device)
+
+processor = AutoProcessor.from_pretrained(model_id)
+
+pipe = pipeline(
+    "automatic-speech-recognition",
+    model=model,
+    tokenizer=processor.tokenizer,
+    feature_extractor=processor.feature_extractor,
+    torch_dtype=torch_dtype,
+    device=device,
+)
+
 
-model = whisper.load_model("large")
 def predict(audio, language):
-    result = model.transcribe(audio=audio, language=language)
+    generate_kwargs = {
+    "task": 'translate',
+    "language": language,
+    "return_timestamps": True,
+    }
+    result = result = pipe(audio, generate_kwargs=generate_kwargs)
     return result["text"]
 
 demo = gr.Interface(