
180 Park Ave - Building 103
Florham Park, NJ
AT&T WATSON (SM) Speech Technologies,
AT&T WATSON (SM) integrates several speech technologies, including speech recognition. Tools allow for tuning recognition, adapting language & acoustic models, and adding custom extensions.
Integrating Incremental Speech Recognition and POMDP-based Dialogue Systems
Iker Arizmendi, Ethan Selfridge, Peter Heeman, Jason Williams
13th annual SIGdial Meeting on Discourse and Dialogue,
2012.
[PDF]
[BIB]
Special Interest Group on Discourse and Dialogue Copyright
The definitive version was published in 2012. , 2012-07-05
{The goal of this paper is to present a first
step toward integrating Incremental Speech
Recognition (ISR) and Partially-Observable
Markov Decision Process (POMDP) based di-
alogue systems. The former provides sup-
port for advanced turn-taking behavior while
the other increases the semantic accuracy of
speech recognition results. We present an In-
cremental Interaction Manager that supports
the use of ISR with strictly turn-based dia-
logue managers. We then show that using
a POMDP-based dialogue manager with ISR
substantially improves the semantic accuracy
of incremental results.
}
Demonstration of AT&T "Let's Go" in the 2010 Spoken Dialog Challenge
Jason Williams, Iker Arizmendi, Alistair Conkie
Proc IEEE Workshop Spoken Language Technology, Berkeley, CA, USA,
IEEE Workshop on Spoken Language Technologies 2010 (SLT),
2010.
[BIB]
{This is a demonstration of the AT&T "Let's Go" bus timetable spoken dialog system entered in the 2010 Spoken Dialog Challenge (http://dialrd.org/sdc). The task in the 2010 Challenge is to provide bus timetable information for a handful of bus lines in Pittsburgh, Pennsylvania. The demonstration consists of the spoken dialog system itself, and an interactive call visualizer. The visualizer enables users to replay past calls, showing the system's output, the ASR input, and the internal belief state. The visualizer can also track current calls in progress, in real time.}