Update app.py

app.py

@@ -9,13 +9,19 @@ UPLOAD_DIR.mkdir(parents=True, exist_ok=True)
 
 _last_segments = []
 
-def process_audio(audio_path):
+def process_audio(audio_path, num_speakers, min_speakers, max_speakers):
     global _last_segments
     if audio_path is None:
         return "⚠️ Please upload an audio file first.", "", ""
+
     try:
         pipeline = get_pipeline()
-        segments = pipeline.process(audio_path)
+        segments = pipeline.process(
+            audio_path,
+            num_speakers=int(num_speakers) if num_speakers else 0,
+            min_speakers=int(min_speakers) if min_speakers else None,
+            max_speakers=int(max_speakers) if max_speakers else None,
+        )
     except Exception as e:
         return f"❌ Error: {e}\n{traceback.format_exc()}", "", ""
 
@@ -35,10 +41,12 @@ def process_audio(audio_path):
     )
     table = f"""<table style='width:100%;border-collapse:collapse;font-size:14px'>
       <thead style='background:#1e3a5f;color:white'>
-        <tr><th style='padding:8px'>Speaker</th><th>Start</th><th>End</th><th style='text-align:left;padding:8px'>Transcript</th></tr>
+        <tr><th style='padding:8px'>Speaker</th><th>Start</th><th>End</th>
+        <th style='text-align:left;padding:8px'>Transcript</th></tr>
       </thead><tbody>{rows}</tbody></table>"""
 
-    return f"✅ Done — {len(segments)} segments", abab, table
+    unique_speakers = len(set(s.speaker for s in segments))
+    return f"✅ Done — {len(segments)} segments, {unique_speakers} speaker(s) detected", abab, table
 
 
 def export_json():
@@ -62,9 +70,26 @@ with gr.Blocks(title="Speech Annotation Pipeline") as demo:
 
     with gr.Row():
         with gr.Column():
-            audio_input = gr.Audio(label="Upload Audio (.wav/.mp3/.flac)", type="filepath")
+            audio_input = gr.Audio(
+                label="Upload Audio (.wav/.mp3/.flac)",
+                type="filepath"
+            )
+
+            gr.Markdown("### 🎯 Speaker Settings")
+            gr.Markdown("*If you know how many speakers, set it below for better accuracy*")
+
+            num_speakers = gr.Number(
+                label="Exact number of speakers (0 = auto-detect)",
+                value=0, precision=0
+            )
+            with gr.Row():
+                min_speakers = gr.Number(label="Min speakers", value=None, precision=0)
+                max_speakers = gr.Number(label="Max speakers", value=None, precision=0)
+
             run_btn = gr.Button("▶ Run Annotation", variant="primary")
             status_box = gr.Textbox(label="Status", interactive=False)
+
+            gr.Markdown("### 📥 Export Results")
             with gr.Row():
                 json_btn = gr.Button("Export JSON")
                 csv_btn  = gr.Button("Export CSV")
@@ -72,12 +97,18 @@ with gr.Blocks(title="Speech Annotation Pipeline") as demo:
             csv_file  = gr.File(label="CSV")
 
         with gr.Column():
-            abab_box   = gr.Textbox(label="Editable Transcript (ABAB)", lines=20,
-                                     placeholder="Annotations will appear here...")
+            abab_box   = gr.Textbox(
+                label="Editable Transcript (ABAB)",
+                lines=20,
+                placeholder="Annotations will appear here..."
+            )
             table_html = gr.HTML(label="Segment Table")
 
-    run_btn.click(fn=process_audio, inputs=[audio_input],
-                  outputs=[status_box, abab_box, table_html])
+    run_btn.click(
+        fn=process_audio,
+        inputs=[audio_input, num_speakers, min_speakers, max_speakers],
+        outputs=[status_box, abab_box, table_html]
+    )
     json_btn.click(fn=export_json, inputs=[], outputs=[json_file])
     csv_btn.click(fn=export_csv,  inputs=[], outputs=[csv_file])