Recent posts

In April 2022, I started as a Visiting Researcher with the Responsible AI (RAI) Human-Centered Technology (HCT) UX Team at Google (Provided by ProUnlimited).

We are excited to see that our paper (Acoustic-phonetic properties of Siri- and human-directed speech) is one of the top 3 downloaded articles from the Journal of Phonetics in the past 90 days!

I’m looking forward to giving a talk Friday, March 25, 2022 for the EU COST (European Cooperation in Science and Technology) Action network for “Language in the Human-Machine Era“.

Speech interactions with voice assistants: a window to human language and cognition Millions of people now regularly interface with technology using spoken language, such as with voice assistants (e.g., Alexa, Siri, Google Assistant). Yet, our scientific understanding of these interactions is still in its infancy. My research explores the cognitive factors shaping speech production, perception, and learning in interactions with voice assistants, using methods from psycholinguistics and acoustic-phonetics.

I’m excited to give a guest lecture on Thursday, November 18th at Boston University for Dr. Kate Lindsey’s Intro. to Linguistics course.

Hey [Siri|Google|Alexa], How do we talk to voice assistants?

For a PDF of slides, please email me directly (mdcohn @ ucdavis . edu)

I was honored to give a keynote today at the inaugural 2021 Human Perspectives on Spoken Human-Machine Interaction (SpoHuMa) FRIAS Junior Researcher Conference, hosted by the Frieberg Institute for Advanced Studies.

See below for slides!

My co-authors, Kris Predeck, Melina Sarian, & Georgia Zellou, and I are thrilled our paper ”Prosodic alignment toward emotionally expressive speech: Comparing human and Alexa model talkers” is now in press in Speech Communication. https://doi.org/10.1016/j.specom.2021.10.003

I’m thrilled that my paper with Georgia Zellou and Bruno Ferenc Segedin has been accepted to the Journal of Phonetics!

The paper examines acoustic-phonetic adjustments when people talk to a Siri vs. a naturally recorded human voice. We find prosodic differences (e.g., increased intensity, smaller f0 range in Siri-DS) as well as some targeted adjustments (vowel hyperarticulation in response to an error made by Siri). Across two experiments varying in error rate, we see differences in the way these register adaptions emerge.

[[Update: Now available online! https://doi.org/10.1016/j.wocn.2021.101123]]

I’m excited to be a keynote speaker at FRIAS Junior Researcher Conference – Human Perspectives on Spoken Human-Machine Interaction this November!

• If you’re interested in submitting a paper to the conference, note the due date of September 3, 2021

Check out my latest Frontiers paper with Georgia Zellou: Prosodic differences in human- and Alexa-directed speech, but similar local intelligibility adjustments

https://www.frontiersin.org/articles/10.3389/fcomm.2021.675704/

I am thrilled to be awarded an ‘Honorable Mention’ for the 2021 UC Davis Chancellor’s Award for Excellence in Mentoring Undergraduate Research (Postdoc/Grad student category). Congrats to all undergraduates, graduate students, postdocs, and faculty honored in the ceremony.

https://urc.ucdavis.edu/chancellors-award-excellence-undergraduate-research