Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
image
image

YAML Metadata Warning: empty or missing yaml metadata in repo card

Check out the documentation for more information.

Assignment 2 – Semantic Video Retrieval

Name:Shivangi Srivastava

UCID: ss5554

Overview

This project builds a semantic retrieval index for video transcript segments. Each transcript segment is converted into a vector embedding and stored in Parquet format for efficient retrieval.

Dataset Schema

Each row contains:

  • video_id: unique video identifier
  • segment_id: segment index
  • start_time: segment start timestamp
  • end_time: segment end timestamp
  • text: transcript text
  • embedding: vector representation of the segment

Methodology

  1. Extract transcript segments
  2. Generate embeddings
  3. Store structured data in Parquet format
  4. Upload dataset to Hugging Face
Downloads last month
52