Update app.py

app.py

@@ -207,6 +207,10 @@ class ParlerTTSStreamer(BaseStreamer):
         else:
             return value
 
+sampling_rate = model.audio_encoder.config.sampling_rate
+frame_rate = model.audio_encoder.config.frame_rate
+
+
 def numpy_to_mp3(audio_array, sampling_rate):
     # Normalize audio_array if it's floating-point
     if np.issubdtype(audio_array.dtype, np.floating):
@@ -230,12 +234,9 @@ def numpy_to_mp3(audio_array, sampling_rate):
     mp3_bytes = mp3_io.getvalue()
     mp3_io.close()
 
-    gr.Info(f"Bytes returned")
+    gr.Info(f"Bytes returned", duration=0.8)
     return mp3_bytes
 
-sampling_rate = model.audio_encoder.config.sampling_rate
-frame_rate = model.audio_encoder.config.frame_rate
-
 @spaces.GPU
 def generate_base(text, description, play_steps_in_s=2.0):
     play_steps = int(frame_rate * play_steps_in_s)
@@ -258,7 +259,7 @@ def generate_base(text, description, play_steps_in_s=2.0):
     thread.start()
 
     for new_audio in streamer:
-        gr.Info(f"Sample of length {round(new_audio.shape[0] / sampling_rate, 2)} seconds ready")
+        gr.Info(f"Sample of length {round(new_audio.shape[0] / sampling_rate, 2)} seconds ready", duration=0.8)
         yield numpy_to_mp3(new_audio, sampling_rate=sampling_rate)