Update app.py

app.py

@@ -1,49 +1,46 @@
 """
 Gradio Web UI for Speech Annotation Pipeline
-Run: python app.py
 """
 
 import json
-import csv
-import io
-import os
 import tempfile
 from pathlib import Path
-
 import gradio as gr
 from pipeline import get_pipeline, to_json, to_csv, to_abab_text
 
 UPLOAD_DIR = Path(tempfile.gettempdir()) / "speech_annotation"
 UPLOAD_DIR.mkdir(parents=True, exist_ok=True)
 
-# ── State ─────────────────────────────────────────────────────────────────────
-_last_segments = []     # module-level cache for export
-
+_last_segments = []
 
-# ── Core processing ───────────────────────────────────────────────────────────
-def process_audio(audio_path: str):
+def process_audio(audio_path):
     global _last_segments
+
     if audio_path is None:
-        return "⚠️ Please upload an audio file.", "", None, None
+        return "⚠️ Please upload an audio file first.", "", "", None
 
     try:
         pipeline = get_pipeline()
         segments = pipeline.process(audio_path)
     except EnvironmentError as e:
-        return f"❌ Config error:\n{e}", "", None, None
+        return f"❌ Config error: {e}", "", "", None
     except Exception as e:
-        return f"❌ Pipeline error:\n{e}", "", None, None
+        import traceback
+        return f"❌ Error: {e}\n{traceback.format_exc()}", "", "", None
 
     if not segments:
-        return "⚠️ No speech detected.", "", None, None
+        return "⚠️ No speech detected.", "", "", None
 
     _last_segments = segments
     abab = to_abab_text(segments)
 
-    # Build HTML table
     rows = "".join(
-        f"<tr><td><b>{s.speaker}</b></td><td>{s.start_fmt}</td><td>{s.end_fmt}</td><td>{s.text}</td></tr>"
-        for s in segments
+        f"<tr style='background:{'#f0f4ff' if i%2==0 else 'white'}'>"
+        f"<td style='padding:6px 8px;font-weight:bold;color:#1e3a5f'>{s.speaker}</td>"
+        f"<td style='padding:6px 8px;color:#555'>{s.start_fmt}</td>"
+        f"<td style='padding:6px 8px;color:#555'>{s.end_fmt}</td>"
+        f"<td style='padding:6px 8px'>{s.text}</td></tr>"
+        for i, s in enumerate(segments)
     )
     table = f"""
     <table style='width:100%;border-collapse:collapse;font-family:monospace;font-size:14px'>
@@ -55,94 +52,69 @@ def process_audio(audio_path: str):
           <th style='padding:8px;text-align:left'>Transcript</th>
         </tr>
       </thead>
-      <tbody>
-        {''.join(f"<tr style='background:{'#f0f4ff' if i%2==0 else 'white'}'>" +
-                 f"<td style='padding:6px 8px;font-weight:bold;color:#1e3a5f'>{s.speaker}</td>" +
-                 f"<td style='padding:6px 8px;color:#555'>{s.start_fmt}</td>" +
-                 f"<td style='padding:6px 8px;color:#555'>{s.end_fmt}</td>" +
-                 f"<td style='padding:6px 8px'>{s.text}</td></tr>"
-                 for i, s in enumerate(segments))}
-      </tbody>
+      <tbody>{rows}</tbody>
     </table>
     """
 
     status = f"✅ Done — {len(segments)} segments detected"
-    return status, abab, table
+    return status, abab, table, None
 
 
 def export_json():
     if not _last_segments:
         return None
-    out = UPLOAD_DIR / "annotation_export.json"
-    to_json(_last_segments, str(out))
-    return str(out)
+    out = str(UPLOAD_DIR / "annotation_export.json")
+    to_json(_last_segments, out)
+    return out
 
 
 def export_csv():
     if not _last_segments:
         return None
-    out = UPLOAD_DIR / "annotation_export.csv"
-    to_csv(_last_segments, str(out))
-    return str(out)
+    out = str(UPLOAD_DIR / "annotation_export.csv")
+    to_csv(_last_segments, out)
+    return out
 
 
-# ── UI ────────────────────────────────────────────────────────────────────────
-CSS = """
-#title { text-align: center; font-size: 28px; font-weight: bold; color: #1e3a5f; margin-bottom: 4px; }
-#subtitle { text-align: center; color: #666; margin-bottom: 20px; }
-.gr-button-primary { background: #1e3a5f !important; }
-"""
+with gr.Blocks(title="Speech Annotation Pipeline") as demo:
 
-with gr.Blocks(css=CSS, title="Speech Annotation Pipeline") as demo:
-
-    gr.HTML("<div id='title'>🎙️ Speech Annotation Pipeline</div>")
-    gr.HTML("<div id='subtitle'>Upload audio → get speaker-labeled transcripts automatically</div>")
+    gr.HTML("<h1 style='text-align:center;color:#1e3a5f'>🎙️ Speech Annotation Pipeline</h1>")
+    gr.HTML("<p style='text-align:center;color:#666'>Upload audio → get speaker-labeled transcripts automatically</p>")
 
     with gr.Row():
         with gr.Column(scale=1):
             audio_input = gr.Audio(
                 label="Upload Audio (.wav / .mp3 / .flac)",
-                type="filepath",
+                type="filepath"
             )
-            run_btn = gr.Button("▶ Run Annotation", variant="primary")
-            status_box = gr.Textbox(label="Status", interactive=False, lines=1)
+            run_btn = gr.Button("▶ Run Annotation", variant="primary", size="lg")
+            status_box = gr.Textbox(label="Status", interactive=False, lines=2)
 
-            gr.Markdown("### 📥 Export Results")
+            gr.Markdown("### Export Results")
             with gr.Row():
                 json_btn = gr.Button("Export JSON")
-                csv_btn = gr.Button("Export CSV")
-            json_file = gr.File(label="JSON Download", visible=True)
-            csv_file = gr.File(label="CSV Download", visible=True)
+                csv_btn  = gr.Button("Export CSV")
+            json_file = gr.File(label="JSON Download")
+            csv_file  = gr.File(label="CSV Download")
 
         with gr.Column(scale=2):
-            gr.Markdown("### 📝 Editable Transcript (ABAB format)")
+            gr.Markdown("### Editable Transcript (ABAB format)")
             abab_box = gr.Textbox(
                 label="Edit transcript below",
                 lines=20,
-                placeholder="Annotations will appear here after processing…",
+                placeholder="Annotations will appear here after processing…"
             )
-            gr.Markdown("### 📊 Segment Table")
+            gr.Markdown("### Segment Table")
             table_html = gr.HTML()
 
-    # Events
+    # Wire up button clicks
     run_btn.click(
         fn=process_audio,
         inputs=[audio_input],
-        outputs=[status_box, abab_box, table_html],
+        outputs=[status_box, abab_box, table_html, json_file],
     )
-
     json_btn.click(fn=export_json, inputs=[], outputs=[json_file])
-    csv_btn.click(fn=export_csv, inputs=[], outputs=[csv_file])
-
-    gr.Markdown(
-        """
-        ---
-        **Tips:**
-        - Set `HF_TOKEN` env var before running (required for speaker diarization)
-        - GPU strongly recommended for files > 5 min
-        - Edits in the transcript box are yours to save — export when ready
-        """
-    )
+    csv_btn.click(fn=export_csv,  inputs=[], outputs=[csv_file])
 
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)