D104: Extracting & Using Gender, Age, Emotion, & Language From Speech

Track D: NEXT BIG THING – Monday 24 April 
14:15 – 15:00

Deep learning, large datasets, and faster computers enable extraction about the characteristics of the speaker, including the speaker language, identity (if they have called before), gender, age, and emotional state with a high level of accuracy from a few seconds of audio recordings. This capability marks the development of a new era in emotionally intelligent dialogue. The presentation concludes with a demonstration and some example use cases.

Presented by: Nagendra Goel