Objective language recovery analysis.
Neurolexi develops clinically grounded neurotechnology for the analysis of language recovery and speech processing in aphasia — measuring speech onset, pronunciation accuracy and naming latency with precision.
Clinical insight, built on neuro-language intelligence
Four principles shape every measurement Neurolexi produces — from the patient app to the therapist dashboard.
Secure & private
Swiss-based hosting and end-to-end data encryption keep patient data protected.
AI-powered
State-of-the-art models for speech recognition and automatic naming-latency analysis.
Neuro-language intelligence
Advanced aNL detection surfaces deeper clinical insight than manual scoring.
Evidence-based
Built on peer-reviewed research and clinical best practice with Swiss institutions.
From a spoken word to an objective result
A picture-naming task on the patient app becomes a clinically meaningful measurement in seconds — recorded, analysed and stored, ready for the therapist.
Patient app
A visual stimulus is shown to the patient on a tablet.
Patient response
The patient speaks the name of the identified object.
Audio upload
The audio waveform and data packet are securely transmitted.
ASR & aNL analysis
Whisper validates the spoken word; the FHNW algorithm detects naming latency.
Result
The validated result is sent back to the patient app.
Secure storage
Results are stored safely in an encrypted PostgreSQL database.
Therapist dashboard
Data becomes available for evaluation, analytics and reporting.
ASR & aNL analysis runs at api.neurolexi.ch using Whisper validation and the FHNW automatic naming-latency algorithm.
Every session, measured and visualised
Naming latency, pronunciation accuracy and overall progress are tracked over time — turning subjective observation into objective, comparable data.
- Naming latency over time — average response time tracked across sessions.
- Pronunciation accuracy — correctness scored automatically per item.
- Session analytics — items practiced, self-corrections and latency improvement.
From research project to certified medical product
The areas below shape how we grow Neurolexi responsibly — combining scientific rigour with the discipline of building a regulated health technology company.
Operations
Lean structures, clear roles and quality-managed processes that scale from spin-off to operating company.
R & D
Continued algorithm research with the FHNW, advancing automatic naming-latency detection and speech analysis.
Go-to-market
Reaching clinics, speech-language therapists and research partners across Switzerland and beyond.
Finance
Sustainable financing, grant strategy and transparent governance to fund long-term clinical impact.
Certification
A clear path toward medical-device conformity (ISO 13485 / MDR) and clinical validation.
Vision
Contributing to the future of digital neurorehabilitation through research-driven neurotechnology.
Grounded in peer-reviewed science
The Neurolexi method is based on clinically tested naming-latency analysis and speech-processing research conducted with Swiss research institutions.
Can an initial phoneme-based algorithm improve automatic naming latency detection during picture naming tasks? A feasibility study
Evaluation of the potential of automatic naming latency detection for different initial phonemes during picture naming task
Automatic detection of naming latency from aphasia patients — using an extended threshold-based method
Patient-friendly speech recognition feedback for aphasia patients
Follow us into the future of health
Are you a clinic, therapist or research partner interested in objective language-recovery analysis? We would love to hear from you.