Update app.py

app.py

@@ -3,6 +3,7 @@ import torch
 from transformers.models.speecht5.number_normalizer import EnglishNumberNormalizer
 from string import punctuation
 import re
+import numpy as np  # Ensure NumPy is imported for audio data processing
 
 from parler_tts import ParlerTTSForConditionalGeneration
 from transformers import AutoTokenizer, AutoFeatureExtractor, set_seed
@@ -62,9 +63,10 @@ def gen_tts(text, description):
             temperature=1.0,
         )
         audio_arr = generation.cpu().numpy().squeeze()
-        return SAMPLE_RATE, audio_arr
+        return SAMPLE_RATE, audio_arr  # Return sample rate and audio array
     except Exception as e:
-        return SAMPLE_RATE, f"Error: {str(e)}"
+        print(f"Error in TTS generation: {str(e)}")
+        return SAMPLE_RATE, np.zeros((SAMPLE_RATE,))  # Return silence in case of error
 
 # Gradio interface
 with gr.Blocks() as block:
@@ -83,7 +85,7 @@ with gr.Blocks() as block:
             audio_out = gr.Audio(label="Parler-TTS generation", type="numpy", elem_id="audio_out")
 
     inputs = [input_text, description]
-    outputs = [audio_out]
+    outputs = audio_out  # Only output the audio component
     run_button.click(fn=gen_tts, inputs=inputs, outputs=outputs, queue=True)
 
 # Launch the interface