Best of: The future of language learning

As kids head back to school this month, we're re-releasing a conversation with cognitive scientist Michael Frank on the future of language learning. Michael studies the similarities and differences in how children and AI systems learn language. He’s working to understand what accounts for the “data gap” between how efficiently children learn language from relatively little input, compared to how much data AI systems need. His goal is to build better scientific models of human language acquisition and to inform how AI systems might learn in more human-like ways. Whether you’re parenting an infant or toddler who is learning to talk or you're simply curious about how language learning unfolds for people across languages and cultures, this one's a great listen. Have a question for Russ? Send it our way in writing or via voice memo, and it might be featured on an upcoming episode. Please introduce yourself, let us know where you're listening from, and share your question. You can send questions to thefutureofeverything@stanford.edu. Episode Reference Links: Stanford Profile: Michael Frank Connect With Us: Episode Transcripts >>> The Future of Everything Website Connect with Russ >>> Threads / Bluesky / Mastodon Connect with School of Engineering >>> Twitter/X / Instagram / LinkedIn / Facebook Chapters: (00:00:00) Introduction Russ Altman introduces this episode with Michael Frank, a professor of biology, psychology and linguistics at Stanford University. (00:02:05) AI and Child Language Learning How AI language learning and child language learning differ. (00:05:23) WordBank and Big Data How large language databases are helping researchers study child language development across cultures. (00:08:04) Learning Across Languages How children learn many different languages through broadly similar developmental processes. (00:08:57) Babbling and Early Speech How babies begin exploring the sounds and signs of their language. (00:10:04) Multimodal Learning How SAYCam & BabyView capture multimodal language learning  (00:13:12) Language as a Social Tool How social cues help children break the code of language. (00:15:12) Learning from Others How children track uncertainty and seek help from social partners. (00:18:05) ManyBabies How a global research consortium studies child development across cultures and communities. (00:20:11) Global Development Research Why large-scale international collaboration is valuable but difficult to fund. (00:21:36) Pragmatics How children learn to infer meaning beyond the literal words people say. (00:23:21) AI, Intent, and Common Sense Why pragmatics matters for building AI systems that understand human intent. (00:24:21) Reading to Children Why books can support language learning through shared attention and rich social interaction. (00:26:38) Early Language Exposure Why it is never too early to interact with babies through language. (00:27:53) Multilingual Children How children can learn multiple languages with few negative consequences. (00:30:35) Conclusion Connect With Us:Episode Transcripts >>> The Future of Everything WebsiteConnect with Russ >>> Threads / Bluesky / MastodonConnect with School of Engineering >>>Twitter/X / Instagram / LinkedIn / Facebook Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.