Latest / Closer Look with Rose Scott

Dissecting the accuracy of voice assistants for minority voices; Mapping project pinpoints where enslaved Blacks lived in Harris County, GA
A new Georgia Tech and Stanford study shows automatic speech recognition (ASR) models, used in voice assistants like Amazon Alexa, may not be as accurate when transcribing English speakers with a minority dialect. However, the study found the transcription of Standard American English (SAE) "significantly outperformed" three dialects: Spanglish, Chicano English and African American Vernacular English. We revisit Rose’s conversation with Camille Harris, PhD candidate in computer science at Georgia Institute of Technology, and lead author of the study. Harris discusses some of the key…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.