Spaces:

AlbertoFH98
/

PodCastena

Running

AlbertoFH98 commited on Jan 27

Commit

a821999

•

1 Parent(s): a8da022

Update utils.py

Files changed (1) hide show

utils.py CHANGED Viewed

@@ -183,7 +183,7 @@ def get_gpt_response(transcription_path, query, logger):
     loader = TextLoader(transcription_path)
     docs = loader.load()
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
     splits = text_splitter.split_documents(docs)
     vectorstore = Chroma.from_documents(documents=splits, embedding=OpenAIEmbeddings())

     loader = TextLoader(transcription_path)
     docs = loader.load()
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=3000, chunk_overlap=250, length_function=len)
     splits = text_splitter.split_documents(docs)
     vectorstore = Chroma.from_documents(documents=splits, embedding=OpenAIEmbeddings())