Additional resources for my paper "Predicting the Micro-Timing of User Input for an Incremental
Spoken Dialogue System that Completes a User's Ongoing Turn" at SigDial 2011
Appendix A: Examples of Shadowing
- source audio
- example 1,
example 2
- aligned recognition result (wavesurfer label file)
- example 1,
example 2
- decision points
- example 1,
example 2
- alignment predictions for next word
- example 1,
example 2
- video of shadowing (recorded externally, including all system delays)*
- example 1
- internal audio of shadowing (potentially omitting sound card delays)†
- example 1
- recording of both audio input and shadow combined†
- example 1
* Breaks/discontinuities in the first few seconds of audio are due
to delays from late initialization of several system components.
The video represents an outdated version of the system.
† Updated audio representing current system state.