LangChainKaltura

A LangChain vector store loader for captions of videos hosted in Kaltura (the basis of UMich's MiVideo service).

Installation

Download the wheel file (.whl) from the latest release on the Releases page.

pip install LangChainKaltura-0.0.1-py3-none-any.whl

Usage

Instantiate KalturaCaptionLoader with all the required parameters, then invoke its load() method…

import os

from LangChainKaltura import \
    KalturaCaptionLoader

captionLoader = KalturaCaptionLoader(
    os.getenv('PARTNERID'),
    os.getenv('APPTOKENID'),
    os.getenv('APPTOKENVALUE'),
    KalturaCaptionLoader.FilterType(
        os.getenv('FILTERTYPE')),
    os.getenv('FILTERVALUE'),
    os.getenv('URLTEMPLATE'))

documents = captionLoader.load()
print(documents)

See the repo for example.py, a more detailed example which reads parameters from .env (based on .env.example) and prints the results as JSON.

Features

(See issue #1 for the most current list of requirements and their completion status.)

• Connecting to Kaltura requires an app token.
• It works only with captioned media, which was presumably written by or approved by media owners. At this time, only SRT captions are supported.
• Captions from media are reorganized into chunks. The chunk duration is configurable, with a default of two minutes.
• It returns a list of LangChain Document object(s), each containing a caption chunk and metadata.
• Caption chunks' metadata contains source URLs to the media, which includes timestamps to the specific chunk of the video.

Test Suite

Run the accompanying tests module to see a complete test of the KalturaCaptionLoader, which includes mocking of the Kaltura API.

python -m tests
python -m unittest tests/test_MiVideoAPI.py

Credits

• Mr. Lance E Sloan (@lsloan) - Development
• Melinda Kraft - Kaltura advising