Get Even More Visitors To Your Blog, Upgrade To A Business Listing >>

Who wants to be the Pittsburgh voice for this open source utility?

---------- Forwarded message ----------
From: Jen Caltrider, Mozilla
Date: Tuesday, July 25, 2017
Subject: I'm sorry, could you repeat that?

A whole new way your Voice matters.
Dear Pittsburgh Yinzers,
We all sound different when we speak. What does that mean for voice recognition? When Siri hears a beautiful Irish brogue or Scottish burr or the accent of a non-native English speaker, she can get tripped up. And Siri has the full force of Apple's voice data collection operation behind her algorithm to help her understand.
What about the small developer who wants to build a voice activated app? Where do they go to get samples of all the beautiful voices speaking from around the world?
Lost in translation?
Up until recently, their options were to spend a fortune buying data to train their algorithms from big corporations or put their product out into the world with an untrained algorithm. It's time for a better option.
Meet Mozilla's new project, Common Voice. It's an open collection of labelled voice data anyone can use to create highly accurate voice recognition software. Well, it will soon be that, with your help. In order to create this valuable public resource, we need people who speak English in all sorts of wonderful ways to go and contribute voice samples. (Note: right now Common Voice is only collecting English samples, but stay tuned, we plan to add other languages very soon.)
Here's what you do — click over to the Common Voice website.
There are two ways you can help. You can click "Speak" and follow the instructions to leave some of your own voice samples. You'll be asked to allow the website to access your microphone. Don't worry, we're Mozilla, we care about your privacy and won't use your microphone for anything but recording the short sentences you'll read. Have fun playing around with that. I know I did.
If leaving a voice recording isn't your thing, you can also just listen. Click the "Listen" link and you'll be asked to listen to some sentences others have read and verify they got it right. That's it. It's actually a ton of fun.
As voice recognition becomes more important in our digital world, everyone — from startups to students at university to that friend of yours who just likes to tinker — should be able to make sure their apps recognize all our beautiful voices. That's how we build a healthy Internet, one step at a time.
Thank you,

Jen Caltrider

Thanks for reading!
You're receiving this email because we think you're neat.

Donate to Mozilla   |   Download Firefox

331 E. Evelyn Avenue Mountain View CA 94041
Legal • Privacy


Mark Rauterkus       [email protected]
Swimming and Water Polo Coach, Schenley High School, Pittsburgh, PA
412 298 3432 = cell

This post first appeared on Mark Rauterkus & Running Mates Ponder Current Even, please read the originial post: here

Share the post

Who wants to be the Pittsburgh voice for this open source utility?


Subscribe to Mark Rauterkus & Running Mates Ponder Current Even

Get updates delivered right to your inbox!

Thank you for your subscription