YouTubeTranscriptApi

[Clffordojuka]

Select Topic Area

Question

Body

I am getting this error when I am trying to test both on the machine and the server"An error occurred: 'YouTubeTranscriptApi' object has no attribute 'get_transcript'"

from youtube_transcript_api import YouTubeTranscriptApi

def fetch_youtube_transcript(video_url: str) -> str:
"""
Fetch YouTube transcript text for the given URL.
"""
if "v=" in video_url:
video_id = video_url.split("v=")[-1].split("&")[0]
else:
video_id = video_url.split("/")[-1]
transcript_data = YouTubeTranscriptApi().get_transcript(video_id, languages=['de', 'en'])
transcript_text = " ".join([entry['text'] for entry in transcript_data])
return transcript_text

[ikrishanaa]

Broo!
You get error because YouTubeTranscriptApi should not be instantiated. Instead of creating an object with YouTubeTranscriptApi(), you should call the method directly on the class.

Change line:
From this-

transcript_data = YouTubeTranscriptApi().get_transcript(video_id, languages=['de', 'en'])

To this

transcript_data = YouTubeTranscriptApi.get_transcript(video_id, languages=['de', 'en'])

Full corrected:

from youtube_transcript_api import YouTubeTranscriptApi

def fetch_youtube_transcript(video_url: str) -> str:
    """
    Fetch YouTube transcript text for the given URL.
    """
    if "v=" in video_url:
        video_id = video_url.split("v=")[-1].split("&")[0]
    else:
        video_id = video_url.split("/")[-1]

    transcript_data = YouTubeTranscriptApi.get_transcript(video_id, languages=['de', 'en'])
    transcript_text = " ".join([entry['text'] for entry in transcript_data])
    return transcript_text

Now it should run without the 'YouTubeTranscriptApi' object has no attribute 'get_transcript' error.

[Clffordojuka]

Imagine trying both calling the method directly from the class and instantiating it, yet still encountering the same error repeatedly. So I don't understand what the problem is

[ebndev]

Hi @Clffordojuka, thanks for being a part of the GitHub Community!

The API & Webhooks category is meant for discussions regarding GitHub's APIs so I will be moving your post to Programming Help, as it may be more relevant there and help someone give you a nudge in the right direction 😄

[Clffordojuka]

Thank I search for both general and this but my eyes couldn't capture either😂

[jihanurrahman33]

Hi @Clffordojuka,

That error happens because you’re instantiating YouTubeTranscriptApi and then calling get_transcript on the instance. In this library, get_transcript is a class method — call it on the class, not on an object.

Also make sure you’re importing the right package (youtube-transcript-api) and that you don’t have a local file named youtube_transcript_api.py shadowing the library.

Fixed & more robust version

from urllib.parse import urlparse, parse_qs
from youtube_transcript_api import YouTubeTranscriptApi, TranscriptsDisabled, NoTranscriptFound, VideoUnavailable


def extract_video_id(url: str) -> str:
    """
    Handles typical YouTube URLs:
    - https://www.youtube.com/watch?v=VIDEO_ID
    - https://youtu.be/VIDEO_ID
    - with extra query params
    """
    parsed = urlparse(url)
    if parsed.hostname in {"www.youtube.com", "youtube.com"}:
        qs = parse_qs(parsed.query)
        if "v" in qs:
            return qs["v"][0]
    # fallback for youtu.be or raw IDs
    return parsed.path.lstrip("/")


def fetch_youtube_transcript(video_url: str) -> str:
    """
    Fetch YouTube transcript text for the given URL.
    Tries German first, then English.
    """
    video_id = extract_video_id(video_url)

    try:
        # ✅ call on the class, NOT an instance
        transcript_data = YouTubeTranscriptApi.get_transcript(
            video_id, languages=["de", "en"]
        )
        return " ".join(entry["text"] for entry in transcript_data)

    except TranscriptsDisabled:
        raise RuntimeError("Subtitles are disabled for this video.")
    except NoTranscriptFound:
        raise RuntimeError("No transcript found for 'de' or 'en'.")
    except VideoUnavailable:
        raise RuntimeError("Video is unavailable or the ID is wrong.")


if __name__ == "__main__":
    print(fetch_youtube_transcript("https://www.youtube.com/watch?v=dQw4w9WgXcQ"))

If it still fails on one machine/server

• Verify you installed the correct lib:

  pip show youtube-transcript-api

• Check you’re not shadowing the package:

  import youtube_transcript_api
  print(youtube_transcript_api.__file__)

  If this points to your project (e.g., .../youtube_transcript_api.py), rename your file.
• Upgrade the lib:

  pip install -U youtube-transcript-api

Hope this helps!

[Clffordojuka]

Okay, let me update and give you feedback

[aarajbhattarai]

I am trying this solution in Kaggle but it's not working. I am also getting the sam error. this worked for meytt_api = YouTubeTranscriptApi() ytt_api.fetch(video_id)

[PRIYAtechky]

The error happens because you’re instantiating YouTubeTranscriptApi as an object, but it’s actually a class with static methods — you should call its methods directly, not via an instance.

from youtube_transcript_api import YouTubeTranscriptApi

def fetch_youtube_transcript(video_url: str) -> str:
    """
    Fetch YouTube transcript text for the given URL.
    """
    if "v=" in video_url:
        video_id = video_url.split("v=")[-1].split("&")[0]
    else:
        video_id = video_url.split("/")[-1]

    # Correct way: call the static method directly
    transcript_data = YouTubeTranscriptApi.get_transcript(video_id, languages=['de', 'en'])

    transcript_text = " ".join([entry['text'] for entry in transcript_data])
    return transcript_text

Change

YouTubeTranscriptApi().get_transcript(...)

to

YouTubeTranscriptApi.get_transcript(...)

[Clffordojuka]

SEE: both can work now, working with the one you suggested, but still getting the error. See my code
import os
from urllib.parse import urlparse, parse_qs
from youtube_transcript_api import (YouTubeTranscriptApi, TranscriptsDisabled, NoTranscriptFound, VideoUnavailable)
import numpy as np
import faiss
from openai import OpenAI
from dotenv import load_dotenv
from typing import Dict, List, Tuple, Any

load_dotenv()

OPENAI_API_KEY validation

if "OPENAI_API_KEY" not in os.environ:
raise ValueError("Please set OPENAI_API_KEY in your environment or .env file")

client = OpenAI()

def extract_video_id(url: str) -> str:
"""
Extract video ID from various YouTube URL formats.
"""
parsed = urlparse(url)
if parsed.hostname in {"www.youtube.com", "youtube.com"}:
qs = parse_qs(parsed.query)
if "v" in qs:
return qs["v"][0]
return parsed.path.lstrip("/")

def fetch_youtube_transcript(video_url: str) -> str:
"""
Fetch YouTube transcript with error handling.
"""
video_id = extract_video_id(video_url)

try:
    transcript_data = YouTubeTranscriptApi.get_transcript(
        video_id, 
        languages=['de', 'en']
    )
    return " ".join(entry['text'] for entry in transcript_data)
except TranscriptsDisabled:
    raise RuntimeError("Subtitles are disabled for this video.")
except NoTranscriptFound:
    raise RuntimeError("No transcript found in English or German.")
except VideoUnavailable:
    raise RuntimeError("Video is unavailable or ID is invalid.")
except Exception as e:
    raise RuntimeError(f"Error fetching transcript: {str(e)}")

def get_openai_embeddings(texts: List[str]) -> np.ndarray:
"""
Generate embeddings using OpenAI's API.
"""
try:
response = client.embeddings.create(
model="text-embedding-ada-002",
input=texts,
)
embeddings_list = [data['embedding'] for data in response.data]
return np.array(embeddings_list).astype('float32')
except Exception as e:
raise RuntimeError(f"Error generating embeddings: {str(e)}")

def chunk_text(text: str, chunk_size: int = 1000, overlap: int = 100) -> List[str]:
"""
Split text into overlapping chunks.
"""
words = text.split()
chunks = []
start = 0

while start < len(words):
    end = start + chunk_size
    chunk = " ".join(words[start:end])
    chunks.append(chunk)
    start += chunk_size - overlap

return chunks

def create_faiss_index(text_chunks: List[str]) -> faiss.Index:
"""
Create FAISS index from text chunks.
"""
embed_vectors = get_openai_embeddings(text_chunks)
dim = embed_vectors.shape[1]
index = faiss.IndexFlatL2(dim)
index.add(embed_vectors)
return index

def similarity_search(
index: faiss.Index,
query: str,
text_chunks: List[str],
k: int = 4
) -> List[str]:
"""
Perform similarity search on FAISS index.
"""
query_embed = get_openai_embeddings([query])
distances, indices = index.search(query_embed, k)
return [text_chunks[i] for i in indices[0]]

def get_answer_from_docs(docs: List[str], question: str) -> str:
"""
Generate answer using GPT-4.
"""
docs_text = " ".join(docs)
prompt = f"""
You are a helpful assistant answering questions about YouTube videos based on the transcript only.

Question: {question}

Transcript excerpts: {docs_text}

Answer in a detailed and factual manner. If not enough information is available, say "I don't know".
"""
try:
response = client.chat.completions.create(
model="gpt-4",
messages=[{"role": "user", "content": prompt}]
)
return response.choices[0].message.content.strip()
except Exception as e:
raise RuntimeError(f"Error generating answer: {str(e)}")

def create_db_from_youtube_video_url(video_url: str) -> Dict[str, Any]:
"""
Create searchable database from YouTube video.
"""
transcript = fetch_youtube_transcript(video_url)
chunks = chunk_text(transcript)
faiss_index = create_faiss_index(chunks)
return {"index": faiss_index, "chunks": chunks}

def get_response_from_query(
db: Dict[str, Any],
query: str,
k: int = 4
) -> Tuple[str, List[str]]:
"""
Search database and generate response.
"""
docs = similarity_search(db["index"], query, db["chunks"], k)
answer = get_answer_from_docs(docs, query)
return answer, docs