A critical assessment of spoken utterance retrieval through approximate lattice representations

Abstract

This paper compares the performance of position-specific posterior lattices (PSPL) and confusion networks applied to spoken utterance retrieval, and tests these recent proposals against several baselines in two disparate domains. These lossy methods provide compact representations that generalize the original segment lattices and provide greater recall and robustness, but have yet to be evaluated against each other in multiple WER conditions for spoken utterance retrieval. Our comparisons suggest that while PSPL and confusion networks have comparable recall, the former is slightly more precise, although its merit appears to be coupled to the assumptions of low-frequency search queries and low-WER environments.

Automatic Summarization

“…The relative performance of lattice and confusion network approaches applied to SUR is evaluated and discussed in [141]. The authors also use a set-of-words approach, which destroys the timing and overlap relationships between the links in the lattice, treating it as a bag-of-words.…”