Update app.py

app.py

@@ -6,13 +6,10 @@ from transformers import AutoTokenizer
 import soundfile as sf
 from pydub import AudioSegment
 import os
-import nltk
-from nltk.data import find
+import re
 from PyPDF2 import PdfReader
 import textwrap
 
-nltk.download('punkt')
-
 # Device configuration
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
@@ -51,10 +48,11 @@ def pdf_to_text(pdf_path):
 tts_model = ParlerTTSForConditionalGeneration.from_pretrained("parler-tts/parler-tts-large-v1").to(device)
 tts_tokenizer = AutoTokenizer.from_pretrained("parler-tts/parler-tts-large-v1")
 
-# Function to split text into sentences
+# Function to split text into sentences using regex
 def split_text_into_sentences(text):
-    sentences = nltk.sent_tokenize(text)
-    return sentences
+    sentence_endings = re.compile(r'[.!?]')
+    sentences = sentence_endings.split(text)
+    return [sentence.strip() for sentence in sentences if sentence.strip()]
 
 # Function to generate audio from text
 def generate_wav_from_text(prompt, description, output_file_prefix):