more ux

app.py

@@ -53,13 +53,18 @@ def diva_audio(audio_input, do_sample=False, temperature=0.001):
 
 def transcribe_wrapper(audio_input, state, model_order):
     spinner = "◒"
+    d_resp = gr.Textbox(
+        value="...",
+        visible=True,
+        label=model_names[0] if not anonymous else f"Model {order}",
+    )
     yield (
         gr.Button(
             value=spinner + " Generating Responses " + spinner,
             interactive=False,
             variant="primary",
         ),
-        "",
+        d_resp,
         state,
     )
 
@@ -121,7 +126,7 @@ def on_page_load(state, model_order):
 def recording_complete(state):
     if state == 1:
         gr.Info(
-            "Once you submit your recording, DiVA will stream back a response! This might take a second."
+            "Once you submit your recording, DiVA will stream back a response! This might take a second as ZeroGPU needs to load model weights into vRAM!."
         )
         state = 2
     return (