docs/integrations/document_loaders/youtube_transcript/

[giscus[bot]]

docs/integrations/document_loaders/youtube_transcript/

YouTube is an online video sharing and social media platform created by Google.

https://python.langchain.com/docs/integrations/document_loaders/youtube_transcript/

[aqib0770]

Currently, YoutubeLoader.from_youtube_url() fails to fetch metadata when the add_video_info parameter is set to True. This is due to changes in YouTube's metadata structure, which pytube relies on.

A more robust alternative is to use yt-dlp instead of pytube for fetching YouTube video metadata.

[raym26]

Currently getting TypeError: 'FetchedTranscriptSnippet' object is not subscriptable when running

loader = YoutubeLoader.from_youtube_url(
"https://www.youtube.com/watch?v=QsYGlZkevEg", add_video_info=False)
loader.load()

[jeremyjousse]

Same for me using UV with Python 3.10, 3.11, 3.12 and 3.13.

[9527andy]

any solution?

[vetri15]

  File "D:\AI\LangChain\langchain_demo\.venv\Lib\site-packages\langchain_community\document_loaders\youtube.py", line 278, in <lambda>
    lambda transcript_piece: transcript_piece["text"].strip(" "),
                             ~~~~~~~~~~~~~~~~^^^^^^^^
TypeError: 'FetchedTranscriptSnippet' object is not subscriptable

I am having the above error , when loading Youtube transcripts. anybody else having the same issue.
It seems the error is on the library itself. while parsing the youtube transcripts.

[9527andy]

TypeError: 'FetchedTranscriptSnippet' object is not subscriptable
same issue, do you fix it?

[vetri15]

TypeError: 'FetchedTranscriptSnippet' object is not subscriptable same issue, do you fix it?

I followed the issue, its been fixed in this git commit . Try upgrading to latest lang chain version , it should go away.

[madhvantyagi]

If anyone knows , why I am getting this problem , let me know what I am doing wrong

Code :

loader = YoutubeLoader.from_youtube_url(
"https://www.youtube.com/watch?v=YOHIViahMlU",
add_video_info=False,
language=["en"], # Specify English language
)

loader.load()

This breaks most of the times , but if i keep running this it works sometimes without even changing anything. I don't know what's wrong

Error:
File "/usr/lib/python3.11/xml/etree/ElementTree.py", line 1728, in close
self.parser.Parse(b"", True) # end of data

ExpatError: no element found: line 1, column 0

During handling of the above exception, another exception occurred:

Traceback (most recent call last):

File "/usr/local/lib/python3.11/dist-packages/IPython/core/interactiveshell.py", line 3553, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)

File "", line 1, in <cell line: 0>
loader.load()

File "/usr/local/lib/python3.11/dist-packages/langchain_community/document_loaders/youtube.py", line 273, in load
transcript_object = transcript.fetch()

File "/usr/local/lib/python3.11/dist-packages/youtube_transcript_api/_transcripts.py", line 134, in fetch
snippets = _TranscriptParser(preserve_formatting=preserve_formatting).parse(

File "/usr/local/lib/python3.11/dist-packages/youtube_transcript_api/_transcripts.py", line 474, in parse
for xml_element in ElementTree.fromstring(raw_data)

File "/usr/local/lib/python3.11/dist-packages/defusedxml/common.py", line 127, in fromstring
return parser.close()

File "/usr/lib/python3.11/xml/etree/ElementTree.py", line 1730, in close
self._raiseerror(v)

File "/usr/lib/python3.11/xml/etree/ElementTree.py", line 1630, in _raiseerror
raise err

File "", line unknown
ParseError: no element found: line 1, column 0

[thinkai101]

this doesnt work anymore, does it? i keep getting ->
HTTPError: HTTP Error 400: Bad Request

[youngnu]

This no longer works. Same Issue occurred
"HTTP Error 400: Bad Request" , "'FetchedTranscriptSnippet' object is not subscriptable"
But I slove this problem. you can access the youtube_transcribe_api directly to get the transcript.