Update app.py

app.py

@@ -42,15 +42,21 @@ def generate_audio(text, model_id):
     num_speakers = pipe_dict["pipe"].model.config.num_speakers
 
     out = []
-    for i in range(min(num_speakers, max_speakers)):
-        forward_params = {"speaker_id": i}
-        output = pipe_dict["pipe"](text, forward_params=forward_params)
-        
-        output =  gr.Audio(value = (output["sampling_rate"], output["audio"].squeeze()), type="numpy", autoplay=False, label=f"Generated Audio - speaker {i}", show_label=True,
-                           visible=True)
+    if num_speakers>1:
+        for i in range(min(num_speakers, max_speakers)):
+            forward_params = {"speaker_id": i}
+            output = pipe_dict["pipe"](text, forward_params=forward_params)
+            
+            output =  gr.Audio(value = (output["sampling_rate"], output["audio"].squeeze()), type="numpy", autoplay=False, label=f"Generated Audio - speaker {i}", show_label=True,
+                               visible=True)
+            out.append(output)
+        out.extend([gr.Audio(visible=False)]*(max_speakers-num_speakers))
+    else:
+        output = pipe_dict["pipe"](text)
+        output =  gr.Audio(value = (output["sampling_rate"], output["audio"].squeeze()), type="numpy", autoplay=False, label="Generated Audio - Mono speaker", show_label=True,
+                               visible=True)
         out.append(output)
-    
-    out.extend([gr.Audio(visible=False)]*(max_speakers-num_speakers))
+        out.extend([gr.Audio(visible=False)]*(max_speakers-1))
     return out