Update app.py

app.py

@@ -27,6 +27,7 @@ from infer.utils_infer import (
     remove_silence_for_generated_wav,
     save_spectrogram,
 )
+from tokenizers import Tokenizer
 from transformers import pipeline
 import click
 import soundfile as sf
@@ -245,7 +246,11 @@ def infer_batch(ref_audio, ref_text, gen_text_batches, exp_name, remove_silence,
     for i, gen_text in enumerate(progress.tqdm(gen_text_batches)):
         # Prepare the text
         text_list = [ref_text + gen_text]
-        final_text_list = convert_char_to_pinyin(text_list)
+        encoding = tokenizer.encode(text_list)
+        tokens = encoding.tokens
+        text_list = ' '.join(map(str, tokens))
+
+        final_text_list = [text_list]
 
         # Calculate reference audio length
         ref_audio_len = audio.shape[-1] // hop_length