AI/ML Engineer, Music Enthusiast.
RESEARCH · INDUSTRY · PROFESSIONAL COMMUNITY
An interpretable, track-aware system ranks likely spectral issues to guide an engineer’s attention. A preliminary single-engineer study reported 89.6 seconds mean time-to-acceptable-mix with assistance; baseline mixing did not reach the stated threshold within the 211-second track. Manuscript submitted to ISMIR 2026.
Exploring how RoomPlan and ARKit geometry, paired with material metadata, can become simulation-ready acoustic environments for physics-based analysis and room-aware audio research.
A proof of concept trained on a small synthetic KARA II dataset to suggest candidate array positions and angles from room dimensions. It is an early ideation aid; coverage and geometry still require proper acoustic modeling and engineering review.
A JUCE plugin prototype that accumulates incoming samples in a fixed FIFO buffer for consistent FFT analysis across DAW buffer settings. Next steps include multi-peak and crest-factor analysis for chord detection, followed by exploring AI-based chord inference.
Acoustic modeling and sound-system design for performance venues and sports facilities, translating coverage, intelligibility, and technical requirements into buildable systems.
Designed and delivered repeatable audio and AV systems across national retail locations, spanning system design, DSP, networked audio, commissioning, and field support.
Experimental Qwen3-ASR encoder LoRA trained with simulated room acoustics and frozen-teacher guidance. Ranked #15 out of 44 models worldwide in Average Word Error Rate (WER) and #5 out of 44 models worldwide in Near Field Speech Recordings on the FFASR leaderboard, powered by Treble Technologies (October 30, 2026).
Master’s projects in data science, AI, and machine learning
Forecast monthly sales for two wine products over the next 12 months, using trend, seasonality, and forecast intervals.
Evaluate a new landing page through an A/B test, comparing subscriber conversion, time on page, and language preference.
Build a retail sales forecasting workflow, from data preparation and model tuning through a deployed prediction interface.
Explore a computer vision workflow for detecting whether workers are wearing helmets in workplace imagery.
Prototype a retrieval-augmented assistant grounded in medical manuals to help users find relevant reference information. Coursework prototype; not for clinical use.
Classify wind turbine generator failures from sensor data to explore proactive maintenance and failure detection.
Compare ensemble classification approaches on provided visa applicant data as a coursework modeling exercise, not an operational decision tool.
Analyze customer data to identify patterns associated with personal loan uptake and develop campaign targeting insights.
Audio and acoustics researcher, AI/ML engineer, and music enthusiast.
At Dolby’s Advanced Technology Group, I work on acoustic mapping, spatial audio, and workflows that connect room capture with acoustic simulation. My background in professional audio engineering informs how I approach research and real-world sound systems.
I’m completing an M.Sc. in Artificial Intelligence and Machine Learning at Montpellier Business School, with coursework in AI, machine learning, deep learning, statistics, time series, and computer vision.
I currently live in the San Francisco Bay Area with my beautiful wife, Grace, and our 2-year-old black Lab, Amber. Outside of audio, I love skiing, hiking, running, playing guitar and banjo, and exploring all the food options the Bay Area has to offer.
Room capture, acoustic simulation, measurement workflows, and validation against physical spaces.
Understanding how room geometry, loudspeaker placement, and acoustic cues shape spatial perception.
Python and C++ prototypes for spectral analysis, audio measurement, and research workflows.
AI and machine learning methods for audio research, with familiarity across AI-based IDEs and developer frameworks.
Front-of-house mixing, audio system setup, signal flow, system tuning, and troubleshooting in live environments.
Acoustic measurement, acoustic mapping, room acoustics, acoustic simulation, room impulse responses, spatial audio, localization, loudspeaker design, coverage modeling, intelligibility, system tuning, validation, live sound, front-of-house mixing, signal flow, DSP, gain structure, commissioning, troubleshooting, Treble SDK, SMAART v8, EASE, Soundvision, Librosa, GCC-PHAT, SRP-PHAT, FOA/HOA ambisonics.
Artificial intelligence, machine learning, statistical learning, deep learning, audio AI, data analysis, exploratory data analysis, statistics, hypothesis testing, A/B testing, time series, classification, regression, decision trees, ensemble methods, neural networks, CNNs, transfer learning, data augmentation, NLP, generative AI, large language models, retrieval-augmented generation, prompt engineering, model evaluation, hyperparameter tuning.
Python, NumPy, SciPy, Pandas, Seaborn, Jupyter, MATLAB (working knowledge), C++, JUCE, FFT, STFT, spectral analysis, audio feature engineering, Streamlit, Flask, Docker, Hugging Face, API development, Git, GitHub, AI-assisted IDEs and developer frameworks.
Swift, SwiftUI, ARKit, RoomPlan, iOS room capture, 3D geometry workflows, room material metadata, and Treble SDK integration.
AV systems integration, Q-SYS Designer, Dante Controller, Dante Virtual Soundcard, Symetrix Composer, BrightSign, Atlona, LEA, M32, M32C, Allen & Heath dLive C3500, IP8, Visionary Solutions, Netgear M4250, UniFi, Wireshark, AVoIP, networked audio, vMix, OBS, NDI.
Revit, AutoCAD, AutoLISP, Bluebeam, SketchUp, system design, rack elevations, one-line drawings, engineering documentation, training, commissioning, and field support.
Spatial audio, room-aware systems, acoustic simulation, localization, and ARKit/RoomPlan-to-simulation workflows within the Advanced Technology Group.
Large-scale sound systems, sports and performance venues, acoustic modeling, Python automation, and engineering validation.
Designed and supported audio systems for retail, hospitality, worship, and live venues, building practical experience in system performance, DSP, and field commissioning.
Graduate study in AI, machine learning, deep learning, statistics, time series, computer vision, optimization, and data analysis. BA in Audio Arts & Acoustics from Columbia College Chicago.
Available for select collaborations, research conversations, and ambitious audio-technology work.
lelandhenry6@gmail.com ↗