Update app.py

app.py

@@ -35,7 +35,10 @@ def asr_transcript(audio):
 
 def real_time_asr(audio, state=""):
     try:
-        audio = np.array(audio)
+        if isinstance(audio, dict) and 'array' in audio:
+            audio = audio['array']
+        else:
+            audio = np.array(audio)
         transcription = asr_transcript(audio)
         state += " " + transcription
         return state, state
@@ -45,7 +48,7 @@ def real_time_asr(audio, state=""):
 # Create the Gradio interface
 iface = gr.Interface(
     fn=real_time_asr,
-    inputs=[gr.Audio(source="microphone", streaming=True), gr.State()],
+    inputs=[gr.Audio(streaming=True), gr.State()],
     outputs=[gr.Textbox(), gr.State()],
     live=True,
     title="Real-Time ASR using Wav2Vec 2.0",