Conceived of, designed, & coded a suite of cloud-based apps to support pronunciation training / accent reduction for non-native speakers of English. Top to bottom, full-stack, micro-services based development (primarily using the MongoDB, Express, React, NodeJS stack with Material UI). Signal processing / explainable AI / machine learning (using Kaldi, TensorFlow, evolutionary optimization,...). Designed/built scalable, cloud-based APIs for pronunciation analysis.Available for consulting in the speech (and language) technology realm -- from simply supporting initial conceptualization / design up to bespoke prototyping of cutting edge solutions. Decades of relevant experience.

Sole responsibility or team lead for design/development of: predictive analytics web services platform; automated related content for video streams; multi-lingual pronunciation prediction (letter-to-sound); co-learning based ASR (automatic speech recognition) resource/model refinement; (adaptive) statistical language modeling; semi-automated production of domain-specific SLMs from web resources; management, production, & acquisition of human language data resources (including management of international contractors); speech dialog based UI; lexical and spoken training resource development and language modeling for an award-winning Chinese ASR system; etc.

Blackboard architectures for knowledge integration; object finding (faces, land mines) in imagery (visible color, hyperspectral);chemical compound classification from MRI spectra; on-line handwriting (incl. cursive) and sketch recognition; etc.