Extract transcripts from Bilibili videos
Fetch video transcripts from Bilibili into LlamaIndex documents.
Why it matters
Users hire this asset to automatically retrieve text transcripts from Bilibili videos for indexing, analysis, or content processing workflows. It eliminates manual transcript copying by programmatically fetching video transcripts through video URLs.
Outcomes
What it gets done
Accept Bilibili video URLs as input parameters
Connect to Bilibili platform using the bilibili_api
Fetch and extract text transcripts from specified videos
Return structured transcript documents for LlamaIndex ingestion
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/li-reader-readers-bilibili | bash Overview
Bilibili Transcript Loader
A LlamaIndex reader that fetches Bilibili video transcripts via the unofficial bilibili_api package. Use for indexing Bilibili video transcripts, accepting the risk of breakage since it relies on an unofficial method.
What it does
The Bilibili Transcript Loader fetches the text transcript of videos from Bilibili, one of the most popular long-form video platforms in China, using the bilibili_api package under the hood. With BilibiliTranscriptReader, users can obtain the transcript of any Bilibili video they specify.
BilibiliTranscriptReader is instantiated with no required configuration, and load_data takes a video_urls list of Bilibili video links to transcribe. The source flags an important caveat: there is no official API for Bilibili transcripts, so the loader depends on the current shape of Bilibili's website, and changes to that site can break it.
When to use - and when NOT to
Use it when you need the spoken content of one or more Bilibili videos as text for LlamaIndex indexing or retrieval. Because it relies on an unofficial method rather than a supported API, expect it to occasionally break when Bilibili changes its site, and treat it as less stable than an official-API-backed loader. Do not build critical, always-on pipelines around it without a plan for handling breakage when Bilibili's website structure changes.
Capabilities
load_data takes a list of Bilibili video URLs and returns each video's text transcript as a LlamaIndex document.
How to install
pip install llama-index-readers-bilibili
Who it's for
Developers who need transcripts of Bilibili videos loaded into LlamaIndex, and can tolerate the loader's dependence on Bilibili's unofficial website structure rather than a supported API.
Source README
Bilibili Transcript Loader
pip install llama-index-readers-bilibili
This loader utilizes the bilibili_api to fetch the text transcript from Bilibili, one of the most beloved long-form video sites in China.
With this BilibiliTranscriptReader, users can easily obtain the transcript of their desired video content on the platform.
Usage
To use this loader, you need to pass in an array of Bilibili video links.
from llama_index.readers.bilibili import BilibiliTranscriptReader
loader = BilibiliTranscriptReader()
documents = loader.load_data(
video_urls=["https://www.bilibili.com/video/BV1yx411L73B/"]
)
Note that there is no official API available for Bilibili Transcript, so changes to the official website can sometimes cause issues.
This loader is designed to be used as a way to load data into Llama Index.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.