Update app.py

app.py

@@ -1,18 +1,38 @@
 import gradio as gr
+import ast
 
 model = gr.Interface.load("huggingface/pyannote/voice-activity-detection")
 
+def format_inference(output):
+    if output:
+        timestamps = []
+        for out in output:
+            timestamps.append(f"Start: {out['start']}s; Stop{out['stop']}s")
+        return "\n".join(timestamps)
+    else:
+        return "No voice activity detected."
+
 def inference(audio_file):
     output = model(audio_file)
-    return output
+    output_list = ast.literal_eval(output)
+    return format_inference(output_list)
 
 inputs = gr.inputs.Audio(label="Input Audio", type="filepath", source="microphone")
-outputs = gr.outputs.Label(type="auto", label = "Voice timestamps")
+outputs = gr.outputs.Textbox(label="Voice timestamps", type="auto")
 title = "Voice Activity Detection"
-description = "Record or upload an audio file and detected human voices will be timestamped."
-article = "<a href = 'pyannote, https://github.com/pyannote/pyannote-audio"
+description = "Record or upload an audio file and detected voices will be timestamped."
+examples = ['samples/talk.wav', 'samples/talk2.wav', 'samples/silence.wav', ]
+article = "pyannote, https://github.com/pyannote/pyannote-audio"
+
+gr.Interface(inference, 
+             inputs, 
+             outputs, 
+             title=title, 
+             description=description, 
+             article=article,
+             examples=examples,
+             theme="grass").launch(debug=True)
 
-gr.Interface(inference, inputs, outputs, title=title, description=description, article=article, theme="grass").launch(debug=True)