Speaker-aware modeling of conversational timing using per-speaker deviation distributions, Markov turn-taking, and unified KDE gap modeling improves alignment with real Switchboard patterns over independence-based baselines.
Improving the naturalness of simulated con- versations for end-to-end neural diarization,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Independence to Interaction: Speaker-Aware Simulation of Multi-Speaker Conversational Timing
Speaker-aware modeling of conversational timing using per-speaker deviation distributions, Markov turn-taking, and unified KDE gap modeling improves alignment with real Switchboard patterns over independence-based baselines.