Listening like a human
Patrick Cormac English
This talk will walk through some of the overarching tensions between use cases in ASR (specific interactions with individuals) and aspects of current state-of-the-art methods used in building and training those systems (one-size-fits-all, big data, generalisation). The discussion will draw on a range of examples across ASR, linguistics, and the intersection thereof to highlight how current systems work, fail, and how they might be designed in the future.
PDT (UTC-7) | 05:20 |
EDT (UTC-4) | 08:20 |
UTC | 12:20 |
Ireland (UTC+1) | 13:20 |
WAT (UTC+1) | 13:20 |
CEST (UTC+2) | 14:20 |
India (UTC+5:30) | 17:50 |