Update app.py

app.py

@@ -186,7 +186,7 @@ def text_to_phonemes(text):
 
 
 @spaces.GPU
-def generate(audio_path, ins, speed, alpha, beta, embedding, steps=100, vocos=False):
+def generate(audio_path, ins, speed, alpha, beta, embedding, steps=200, vocos=False):
     ref_s = other_tts.compute_style(audio_path)
     print(ref_s.size())
     s_prev = None
@@ -199,7 +199,7 @@ def generate(audio_path, ins, speed, alpha, beta, embedding, steps=100, vocos=Fa
         synthaud, s_prev = other_tts.long_inference_segment(i, diffusion_steps=steps,
                                                             alpha=alpha, beta=beta, is_phonemes=True,
                                                             embedding_scale=embedding, prev_s=s_prev, ref_s=ref_s,
-                                                            speed=speed, t=0.8)
+                                                            speed=speed, t=0.8, vocos=vocos)
 
         n_trim = int(len(synthaud) * 0.009)
         synthaud[:n_trim] = 0