Voicebench

Which speech model hears real people best, and fastest?

Take part

Add your voice: read the script below into your phone. It takes a couple of minutes. Your recording is published with the results, without your name or anything else about you.

How to record

  1. Find a quiet room, open Voice Memos (iPhone) or Recorder (Android), press record, and hold the phone about 2-3 inches from your mouth, with the bottom of the phone closest to your mouth.
  2. Keep it there the whole time.
  3. Wait two seconds, then read the lines below at your normal pace, pausing briefly after each one.
  4. If you stumble, just keep going.

What to read

  1. Good morning.
  2. What's the weather like in London today?
  3. And what about Sunday?
  4. Set a timer for ten minutes.
  5. Cancel that.
  6. Wake me up at six tomorrow.
  7. Send a message to David saying I'm running about fifteen minutes late, so please start without me and I'll join you soon.
  8. Yes, send it.
  9. Remind me to buy eggs, strawberries, and orange juice when I leave work this evening.
  10. Play some quiet jazz in the kitchen.
  11. A little louder, please.
  12. Stop.
  13. Show me the fastest way to the airport from here, and tell me whether there's heavy traffic on the bridge right now.
  14. Did anyone call while I was out?
  15. Schedule a meeting with Emma for next Wednesday at three, and move my usual team lunch to Thursday.
  16. Search for a vegetable soup recipe that serves four.
  17. No, the other one.
  18. How do you say thank you in Spanish?
  19. Can you explain how a rainbow forms, using simple words that a young child could easily understand?
  20. Thanks, that's all.

When you're done

Send me the voice recording back when you're done! Thanks for your help!

Not in touch with us already? Send your recording as an issue on GitHub: the issue opens ready to fill in, and you attach the file there. An issue is public and shows your GitHub username; the published recording and the results never do.