Update app.py

app.py

@@ -1,3 +1,24 @@
+import gradio as gr
+from nemo.collections.asr.models import ASRModel
+
+# Load the NeMo ASR model
+model = ASRModel.from_pretrained("nvidia/canary-1b")
+model.eval()
+
+def transcribe(audio):
+    if audio is None:
+        raise gr.InterfaceError("Please provide some input audio: either upload an audio file or use the microphone")
+
+    # Perform speech recognition
+    transcription = model.transcribe([audio])
+
+    return transcription[0]
+
+audio_input = gr.components.Audio()
+
+iface = gr.Interface(transcribe, audio_input, "text", title="ASR with NeMo Canary Model")
+iface.launch()
+
 '''
 import gradio as gr
 from transformers import pipeline
@@ -5,7 +26,7 @@ from transformers import pipeline
 asr_pipeline = pipeline("automatic-speech-recognition", model="nvidia/canary-1b", device=0)
 qa_pipeline = pipeline("question-answering", model="LLAMA/llama3-base-qa", tokenizer="LLAMA/llama3-base-qa")
 tts_pipeline = pipeline("text-to-speech", model="patrickvonplaten/vits-large", device=0)
-'''
+
 
 import gradio as gr
 import json
@@ -187,7 +208,7 @@ with gr.Blocks(
 demo.queue()
 demo.launch()
 
-'''
+
 
 
 # Function to capture audio using Canary ASR