Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025ACM Transactions on Asian and Low-Resource Language Information ProcessingOpen Access

Abstractive Summarization for Urdu Video Description Generation

View Full Paper
Ask AI
Bookmark
Share

Authors

AFAli FaheemFUFaizad UllahMAMuhammad Sohaib Ayub

Discussion

Loading...

Member takes

Overview

This research demonstrates the creation of Urdu video transcription datasets, suggesting improvements in Urdu content accessibility.

Key Points

  • The proposed method generates coherent Urdu video descriptions, enhancing viewer engagement on social media platforms.
  • Leveraging transfer learning with the mT5 model, the research employs ROUGE scores for evaluating the generated summaries.
  • Evaluations show that the generated descriptions are more accurate compared to translated ground truth, validating the approach.
  • This work establishes a foundation for further advancements in multilingual content generation and low-resource languages.

Cite This Study

Faheem et al. (2025) studied this question.

synapsesocial.com/papers/68c1d5f754b1d3bfb60f8ebdhttps://doi.org/10.1145/3762992
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Headline generation based on statistical translation2000 · 254 citations
  2. 2A Neural Attention Model for Abstractive Sentence Summarization2015 · 700 citations
  3. 3FFCI: A Framework for Interpretable Automatic Evaluation of Summarization2022 · 16 citations