added alternatives,

app.py

@@ -46,7 +46,7 @@ def calculateEmbeddings(sentences,tokenizer,model):
 # explicit no operation hash function, because model and tokenizer are not going to change
 @st.cache(hash_funcs={transformers.models.bert.tokenization_bert_fast.BertTokenizerFast: lambda _: None, transformers.models.bert.modeling_bert.BertModel: lambda _: None})
 def load_model_and_tokenizer():
-    multilingual_checkpoint = 'sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2'
+    multilingual_checkpoint = 'sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2'  #alternative: SZTAKI-HLT/hubert-base-cc 
     tokenizer = AutoTokenizer.from_pretrained(multilingual_checkpoint)
     model = AutoModel.from_pretrained(multilingual_checkpoint)
     print(type(tokenizer))
@@ -58,7 +58,7 @@ model,tokenizer = load_model_and_tokenizer();
 raw_text_file = 'joint_text_filtered.md'
 all_sentences = load_raw_sentences(raw_text_file)
 
-embeddings_file = 'multibert_embedded.pt'
+embeddings_file = 'multibert_embedded.pt' #alternative: hunbert_embedded.pt
 all_embeddings = load_embeddings(embeddings_file)
 
 
@@ -68,7 +68,7 @@ st.caption('[HU] Adjon meg egy tetszőleges kifejezést és a rendszer visszaadj
 
 
 
-text_area_input_query = st.text_area('[HU] Beviteli mező - [EN] Query input',value='Mikor van a leadási hataridő?')
+text_area_input_query = st.text_area('[HU] Beviteli mező - [EN] Query input',value='Mik a részfeladatok?')
 
 if text_area_input_query:
     query_embedding = calculateEmbeddings([text_area_input_query],tokenizer,model)