YouTube Transcript API

Warnings

• breaking The static methods `YouTubeTranscriptApi.get_transcript`, `YouTubeTranscriptApi.get_transcripts`, and `YouTubeTranscriptApi.list_transcripts` were removed in `v1.2.0`. These methods were previously deprecated in `v1.0.0`.
  Fix: Instead of calling static methods, instantiate the `YouTubeTranscriptApi` object and use its instance methods like `ytt_api.fetch(video_id)` or `ytt_api.list_transcripts(video_id)`.
• breaking In `v1.1.0`, the library refactored caption retrieval from scraping the `/watch` HTML to fetching from the Innertube API. This change initially did not support authentication, potentially causing issues for users relying on authenticated requests.
  Fix: Ensure your environment and usage patterns are compatible with the Innertube API approach. Monitor for `PoTokenRequired` exceptions, which may indicate new authentication requirements, and report them to the library maintainers.
• gotcha YouTube frequently blocks requests from known cloud provider IPs or if a high volume of requests originates from a single IP. This can lead to `IpBlocked` exceptions.
  Fix: For heavy usage or deployment on cloud platforms, consider using rotating proxy services (e.g., Webshare) configured with the `WebshareProxyConfig` class when initializing `YouTubeTranscriptApi`.
• gotcha Ensure you pass only the YouTube video ID, not the full video URL, to the API methods. For example, for `https://www.youtube.com/watch?v=VIDEO_ID`, the video ID is `VIDEO_ID`.
  Fix: Extract the video ID from the URL before passing it to `fetch` or `list_transcripts`.
• gotcha If you encounter `NoTranscriptFound` or `TranscriptsDisabled` exceptions, it means the video either genuinely lacks captions or has them explicitly turned off by the uploader. Some automatically generated captions may also not be available.
  Fix: Handle these exceptions gracefully in your code. You can use `list_transcripts` to check for available languages before attempting to fetch a specific one.

Install

• pip install youtube-transcript-api Install latest version

Imports

• YouTubeTranscriptApi

  from youtube_transcript_api import YouTubeTranscriptApi

  As of v1.2.0, the static methods `get_transcript`, `get_transcripts`, and `list_transcripts` were removed. Instantiate `YouTubeTranscriptApi` and use instance methods like `fetch`.
• NoTranscriptFound

  from youtube_transcript_api import NoTranscriptFound

  This exception is raised if no transcript can be found for the given video ID.
• TranscriptsDisabled

  from youtube_transcript_api import TranscriptsDisabled

  This exception is raised if transcripts are explicitly disabled for the video.
• NoManualTranscriptFound

  from youtube_transcript_api import NoManualTranscriptFound

  This exception is raised if no manually created transcript is available for the video in the requested language(s).
• IpBlocked

  from youtube_transcript_api import IpBlocked

  This exception indicates that the request was blocked, often due to aggressive rate limiting or IP blocking by YouTube, especially from cloud providers.

Quickstart

This quickstart demonstrates how to fetch a transcript for a given YouTube video ID using the `YouTubeTranscriptApi` client. It includes error handling for common issues like missing or disabled transcripts and IP blocking. The example retrieves the 'Never Gonna Give You Up' transcript and prints its beginning.

from youtube_transcript_api import YouTubeTranscriptApi, NoTranscriptFound, TranscriptsDisabled, IpBlocked

video_id = "_dQ8wY76uI8" # Example: 'Never Gonna Give You Up' by Rick Astley

try:
    # Instantiate the API client
    ytt_api = YouTubeTranscriptApi()

    # Fetch the transcript for the video
    # By default, it attempts to get English. You can specify languages=['de', 'en'] for priority.
    transcript = ytt_api.fetch(video_id)

    # Print the full transcript text
    full_text = " ".join([item['text'] for item in transcript])
    print(f"Transcript for video ID {video_id}:\n{full_text[:500]}...")

    # You can also list available transcripts and their languages
    # transcript_list = ytt_api.list_transcripts(video_id)
    # for t in transcript_list:
    #     print(f"Available transcript: {t.language} ({'generated' if t.is_generated else 'manual'})")

except NoTranscriptFound:
    print(f"No transcript found for video ID: {video_id}")
except TranscriptsDisabled:
    print(f"Transcripts are disabled for video ID: {video_id}")
except IpBlocked:
    print(f"Your IP was blocked by YouTube. Consider using a proxy.")
except Exception as e:
    print(f"An unexpected error occurred: {e}")