YouTube Transcript API for LangChain

import httpx
from langchain.schema import Document
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_community.vectorstores import Chroma
from langchain.chains import RetrievalQA

# 1. Fetch transcript from YouTubeTranscripts.co
def load_youtube_transcript(video_url: str, api_key: str) -> Document:
    response = httpx.get(
        "https://api.youtubetranscripts.co/v1/transcript",
        params={"url": video_url, "format": "text"},
        headers={"x-api-key": api_key},
    )
    response.raise_for_status()
    data = response.json()
    return Document(
        page_content=data["text"],
        metadata={"title": data["title"], "channel": data["channel"], "source": video_url},
    )

# 2. Load and split the transcript
doc = load_youtube_transcript(
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
    api_key="YOUR_API_KEY",
)

splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
chunks = splitter.split_documents([doc])

# 3. Create vector store and retrieval chain
vectorstore = Chroma.from_documents(chunks, OpenAIEmbeddings())
qa_chain = RetrievalQA.from_chain_type(
    llm=ChatOpenAI(model="gpt-4o"),
    retriever=vectorstore.as_retriever(),
)

# 4. Ask questions about the video
answer = qa_chain.invoke("What are the main topics discussed?")
print(answer["result"])