This Day in AI Podcast

EP38: Ed Sheeran Listens to Our Podcast, Deep Fakes & Frontier Risks and AI Ears: SALMONN Model

10 snips
Oct 27, 2023
Guest
Ed Sheeran
Ed Sheeran, a famous musician, makes a surprise appearance and discusses his love for the podcast. The podcast also covers topics such as deep fakes and their potential dangers, AI-generated voices becoming undetectable, challenges in web crawling, limitations of current PDF to text technology, and the idea of creating an agent as a moral conscious.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
ADVICE

Always Verify Media Before Sharing

  • Verify surprising audio/video claims before sharing because even believable clips can be faked and weaponized.
  • Check sources and contextual cues rather than assuming authenticity from realistic audio or visuals.
INSIGHT

SALMONN Adds 'Hearing' To Multimodal AI

  • ByteDance's SALMONN model analyzes all audio types (voices, music, background) to infer context, emotion, and environment.
  • Integrating SALMONN with vision and text models fills a major sensing gap for richer multimodal understanding.
ADVICE

Combine Senses For Better Video Understanding

  • Use combined audio, vision, and text embeddings to extract richer meaning from videos for editing, chaptering, and sentiment analysis.
  • Apply models that analyze delivery, background sounds, and frames, not just transcripts, to approximate human editing decisions.
Get the Snipd Podcast app to discover more snips from this episode
Get the app