Closed rrolfe closed 5 years ago
Hi there! Thank you for taking the time to submit a talk! Speakers like you make the DC tech community awesome — and we’re glad you’re here.
Rest assured that organizers of quite a few meetups have just been notified of your proposal.
Want to target one or two groups specifically? Mention the group by tagging @dctech/[group]
— but please avoid tagging groups at random or tagging more than one or two. Thanks for keeping this group productive and spam-free!
Not interested in giving this talk anymore? Not a problem at all! Thanks for considering it in the first place! Go ahead and close this issue, so we know not to bother you about it.
About You
Your Name: Rebecca Rolfe
Twitter handle (optional): N/A, but my site is rebeccarolfe.com
The best way to reach out to you: Email at rrolfe@npr.org
A quick bio: Rebecca worked as a production designer at CNN.com before trying out the west coast tech life at Google. She now focuses on the future of voice assistants at NPR in Washington, DC. She studied visual communication at UNC-Chapel Hill and has a masters in Digital Media from Georgia Tech.
Your Talk
Title: Radio hosts are the voice assistant personas of the future
What your talk is about: Voice designers are told personas are an absolute must for good voice design. If you leave it up to chance, it fractures the experience. But for its early voice projects, NPR didn't have a persona in mind. There's no 'one voice' on NPR; we didn't want to mandate one for voice platforms. That works for awhile, but not forever.
In 2019, NPR re-evaluated its voice flows and conducted research with listeners on preferred phrasing, pitch, and speed of digitized voices. We talked with hosts and considered what voices are missing on air and how to bring that to voice assistants. We will share before and after flows and the process used to help us move closer towards a voice assistant tone that 'sounds' like NPR (without having to build a complex backstory.)
How long will your talk be?
Meta
Do you need help crafting your talk?