How Voice AI will change the world

Computers are starting to sound more human than ever. Recent progress in audio AI and LLMs is making conversations with machines feel like a natural extension of how we already communicate.

This week’s episode of Babbage, a science and technology podcast from The Economist, looks at these changes. In it, our co-founder Mati Staniszewski joins Alex Hern (The Guardian), Vasco Pedro (Unbabel), and Steven Johnson (Google Labs) to talk about how these systems work and what they mean for the future.

For The Economist subscribers:

The episode highlights something we’ve long believed: the human voice makes technology feel familiar, even across different languages and formats. When machines sound natural, they bring knowledge and stories to more people in ways that feel intuitive and accessible. Voice AI is already helping teachers reach students in their native languages and giving people with disabilities tools to express themselves. It’s also creating new ways to share stories and produce richer media.

Because its potential is broad, audio AI needs safeguards. That’s why we’ve focused on tools that let people verify AI-generated content and ensure it’s used responsibly. Our goal is to make Voice AI a tool people can rely on—a new way to interact with the digital world, on human terms.