ZeroOneEta is a specialist signal-processing studio — microphone-array beamforming to neural inference, grounded in IEEE-published research and shipped to production.
// about
ZeroOneEta is a specialist signal-processing studio. We focus on four things: digital signal processing, acoustics, beamforming, and deep learning — from microphone-array geometry and room-impulse modelling to neural networks trained on raw signals.
Our work is grounded in peer-reviewed research — a PhD in signal processing and acoustics, with published, independently cited work in IEEE, EUSIPCO, and EURASIP venues. Every engagement begins at the mathematics and ends with working, deployed code — not just a notebook.
We work remotely with research labs, deep-tech startups, and enterprises worldwide — from focused technical spikes to multi-month engagements. CET timezone, 24-hour response time.
// services
Four deep specialisms — not a long generic menu. Remote-first, fixed-scope contracts or ongoing advisory, available globally from a CET timezone overlapping European, Middle Eastern, and US Eastern hours.
Spectral analysis, FFT pipelines, FIR/IIR filter design, convolution, adaptive filtering, noise reduction, and feature extraction — from the maths to optimised, real-time code.
Room impulse-response and reverberation modelling, sound-field analysis, source localisation, and acoustic simulation for audio, speech, and sensing applications.
Microphone- and sensor-array design, delay-and-sum and MVDR beamformers, direction-of-arrival estimation, and spatial filtering for noisy, multi-source environments.
Neural networks for signals and audio — CNNs, RNNs, and Transformers on raw waveforms and spectrograms. Classification, denoising, and model deployment to production.
// portfolio
Research is only half the job. We ship working audio and DSP software — interactive tools we've built and deployed, turning the theory above into systems people actually use.
Interactive GUI for simulating room impulse responses with pyroomacoustics. Configure geometry, place microphone arrays, and visualise RIR, RT60, and acoustic properties.
Browser-based audio annotation tool for labelling speech, environmental sounds, and signal events — building the clean, training-ready datasets deep-learning models depend on.
Pipeline for generating and augmenting audio datasets — adding controlled noise, reverberation, and room simulation to train robust signal and speech models.
// who we work with
Our work sits where peer-reviewed acoustics meets shipping product. These are the kinds of R&D teams and deep-tech startups we're a strong fit for — each domain backed by published, independently cited research.
Multi-microphone beamforming, acoustic echo estimation, and DOA estimation for hearing devices and smart earbuds — extracting a target voice from noise.
Far-field speech enhancement, ego-noise suppression, and deep-learning denoising — the front-end signal chain that makes voice interfaces robust in the real world.
Acoustic mapping and SLAM from echoes, reflector localisation, and audio-based perception for platforms that need to sense space through sound.
Room impulse-response modelling, reverberation, and acoustic reflector estimation for realistic spatial audio in AR/VR and immersive experiences.
Model-based localisation, distance estimation, and array signal processing for GPS-denied indoor navigation and acoustic sensing systems.
Microphone-array design, adaptive filtering, and DSP/deep-learning pipelines for speakers, conferencing hardware, and professional audio products.
// stack & availability
Everything you need to know before starting a project together.
// contact
Describe what you're trying to build, the problem domain, and your timeline. We'll respond within 24 hours with an honest assessment of fit and a rough scope.
No obligation. No sales call required for an initial technical assessment.