Nathan Bodenstab

Nathan Bodenstab
research category image

Main area of research
Speech recognition

Nate Bodenstab is a Research Scientist Manager and joined Nuance in 2007. He received his Ph.D. in Computer Science from OGI/OHSU in 2012, where his thesis centered around efficient constituent parsing. He currently leads a team of researchers focused on language modeling for virtual assistants.

Selected articles

Multi-Pass Pronunciation Adaptation

A mapping between words and pronunciations (potential phonetic realizations) is a key component of speech recognition systems. Traditionally, this has been encoded in a lexicon where each pronunciation

Read more

Finite-State Chart Constraints for Reduced Complexity Context-Free Parsing Pipelines

We present methods for reducing the worst-case and typical-case complexity of a context-free parsing pipeline via hard constraints derived from finite-state pre-processing. We perform O(n) predictions

Read more

Efficient matrix-encoded grammars and low latency parallelization strategies for CYK

We present a matrix encoding of context free grammars, motivated by hardware-level efficiency considerations. We find efficiency gains of 2.5–9 for exhaustive inference and approximately

Read more

Unary constraints for efficient context-free parsing

We present a novel pruning method for context-free parsing that increases efficiency by disallowing phrase-level unary productions in CKY chart cells spanning a single word.

Read more

Beam-width prediction for efficient context-free parsing

Efficient decoding for syntactic parsing has become a necessary research area as statistical grammars grow in accuracy and size and as more NLP applications leverage

Read more


1 2