Project information

  • Category: Deep Learning, Natural Language Processing
  • Role: Solo Researcher
  • Project date: November - December, 2022
  • Project URL: Paper

Using AI to explain AI

Leveraged state-of-the-art (SOTA) speech-to-text transcription models and transformer architectures to deliver clear and concise explanations on complex AI topics.

Developed a Semantic Search Engine utilizing FAISS to recommend relevant podcast episodes from the Lex Fridman Podcast, enhancing the user experience for listeners.

Utilized transformer model architectures such as BART, Pegasus, and T5 for abstractive summarization, producing highly accurate and succinct summaries.