agregar download de tokenizador

app.py

@@ -4,6 +4,8 @@ from nltk.tokenize import word_tokenize
 import gradio as gr
 
 
+nltk.download('punkt')
+
 # Use gensim Keyvectors to read the embbedings 
 wordvectors_file_vec = 'smaller_model_spa.txt'
 smaller_model = KeyedVectors.load_word2vec_format(wordvectors_file_vec)
@@ -12,7 +14,7 @@ with open('stop_words.pkl', 'rb') as f:
 
 
 def filter_words(x):
-    word_tokens = word_tokenize(x)
+    word_tokens = word_tokenize(x, language = "spanish")
     filtered_sentence = [w for w in word_tokens if not w.lower() in stop_words]
     return filtered_sentence