Update app.py

app.py

@@ -113,7 +113,38 @@ def transcribe(audio_filepath, src_lang, tgt_lang, pnc):
 
 	return output_text
 
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+torch.random.manual_seed(0)
+
+model = AutoModelForCausalLM.from_pretrained(
+    "microsoft/Phi-3-mini-128k-instruct", 
+    device_map="auto", 
+    torch_dtype="auto", 
+    trust_remote_code=True, 
+)
+tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-128k-instruct")
+
+messages = [
+    {"role": "user", "content": str(output_text)},
+]
+
+pipe = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+)
+
+generation_args = {
+    "max_new_tokens": 500,
+    "return_full_text": True,
+    "temperature": 0.0,
+    "do_sample": False,
+}
 
+output_text = pipe(messages, **generation_args)
+print(output[0]['generated_text'])
     
 	
 with gr.Blocks(
@@ -128,15 +159,13 @@ with gr.Blocks(
 	theme=gr.themes.Default(text_size=gr.themes.sizes.text_lg) # make text slightly bigger (default is text_md )
 ) as demo:
 
-	gr.HTML("<h1 style='text-align: center'>NeMo Canary model: Transcribe & Translate audio</h1>")
+	gr.HTML("<h1 style='text-align: center'>Your amazing AI assistant</h1>")
 
 	with gr.Row():
 		with gr.Column():
 			gr.HTML(
-				"<p><b>Step 1:</b> Upload an audio file or record with your microphone.</p>"
-
+				"<p><b>Step 1:</b> Record with your microphone.</p>"
             )
-
 			audio_file = gr.Audio(sources=["microphone"], type="filepath")
 
 
@@ -151,7 +180,7 @@ with gr.Blocks(
 				label="Model Output",
 				elem_id="model_output_text_box",
 			)
-
+    
 	go_button.click(
 		fn=transcribe, 
 		inputs = [audio_file],