Update app.py

app.py

@@ -35,6 +35,9 @@ def parse_multilingual_text(input_text):
 
 @spaces.GPU
 def generate_segment_audio(text, lang, speaker_url, pipe):
+    if not isinstance(text, str):
+        text = text.decode("utf-8") if isinstance(text, bytes) else str(text)
+
     stoks = pipe.t2s.generate([text], lang=[lang])
     audio_data = pipe.generate(stoks, speaker_url, lang)
     resample_audio = resampler(newsr=24000)
@@ -42,6 +45,7 @@ def generate_segment_audio(text, lang, speaker_url, pipe):
     audio_np = audio_data_resampled.cpu().numpy()
     return audio_np
 
+
 # Function to concatenate audio segments
 def concatenate_audio_segments(segments):
     concatenated_audio = np.concatenate(segments, axis=0)