Would You Lend Your Voice To Our Experiment?

Mhari Shaw

We need your help — and your voice.

From a robot in your phone to a smart speaker in your kitchen, voice-to-text algorithms are moving into more and more aspects of our lives.

But how well do they understand English speakers of all backgrounds? We're running an experiment to find out.

This is where you come in: Record yourself speaking (we'll give you prompts), send us the clips and we'll have the machines interpret them. It doesn't matter if you know of or have used the devices or services we're testing. If you can speak, we'd like to hear from you. Here's the link to the experiment.

Thanks!

Copyright 2024 NPR. To see more, visit https://www.npr.org.

  • Facebook
  • Twitter
  • LinkedIn
Huo Jingnan (she/her) is an assistant producer on NPR's investigations team. She helps with reporting, research, and production both on the team and in the network. She was the primary data reporter on Coal's Deadly Dust, a project investigating black lung disease's resurgence. The project won an Edward Murrow Award and NASEM Communications award, and was nominated for a George Foster Peabody award.