Update app.py

app.py

@@ -4,7 +4,7 @@ import numpy as np
 
 asr_model = "distil-whisper/distil-medium.en"
 
-transcriber = pipeline("automatic-speech-recognition", model=asr_model)
+asr_pipe = pipeline("automatic-speech-recognition", model=asr_model)
 
 def transcribe(stream, new_chunk):
     sr, y = new_chunk
@@ -15,13 +15,24 @@ def transcribe(stream, new_chunk):
         stream = np.concatenate([stream, y])
     else:
         stream = y
-    return stream, transcriber({"sampling_rate": sr, "raw": stream})["text"]
+    return stream, asr_pipe({"sampling_rate": sr, "raw": stream})["text"]
 
+demo = gr.Blocks()
 
-demo = gr.Interface(
-    transcribe,
-    ["state", gr.Audio(sources=["microphone"], streaming=True)],
-    ["state", "text"],
+
+mic = gr.Interface(
+    fn = transcribe,
+    inputs = [
+        "state", gr.Audio(sources=["microphone"], streaming=True)],
+    outputs = ["state", "text"],
+    layout="horizontal",
+    theme="huggingface",
+    title="Whisper & BERT demo - Intent Classification",
+    description=(
+        "Transcribe audio inputs with Whisper ASR model and detect intention from the text. Use BERT NLP model to classify the intention 
+            "as one of the commands to command a light."
+    ),
+    allow_flagging="never",
     live=True,
 )