Update app.py

app.py

@@ -2,7 +2,6 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 import torch
 import soundfile as sf
 from xcodec2.modeling_xcodec2 import XCodec2Model
-from IPython import display
 import torchaudio
 import spaces
 import gradio as gr
@@ -59,6 +58,9 @@ def infer(sample_audio_path, target_text):
     with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as f:
 
         waveform, sample_rate = torchaudio.load(sample_audio_path)
+        if len(waveform[0])/sample_rate > 15:
+            gr.Warning("Trimming audio to first 15secs.")
+            waveform = waveform[:, :sample_rate*15]
 
         # Check if the audio is stereo (i.e., has more than one channel)
         if waveform.size(0) > 1: