# NeuroGram — Transforming Thoughts into Words

> A non-invasive, AI-powered brain-computer interface restoring communication for individuals with severe speech impairments.

**Canonical URL:** https://www.nglab.tech/  
**Last updated:** 2026-04-16

---

## What is NeuroGram?

NeuroGram is a next-generation brain-computer interface (BCI) that uses non-invasive EEG technology combined with deep learning to decode neural signals and convert them into text or synthesised speech — in real time.

It is designed for individuals who have lost the ability to speak due to conditions such as ALS (Amyotrophic Lateral Sclerosis), locked-in syndrome, stroke, or severe cerebral palsy.

---

## The Problem

Over 5 million people globally are non-verbal due to neurological conditions. Existing communication solutions are slow (1–5 words per minute), expensive, and require extensive calibration per user. NeuroGram aims to change that.

---

## How It Works

1. **Signal Acquisition** — Non-invasive EEG cap records neural activity with 256 dry electrodes.
2. **Preprocessing** — Artefact rejection and signal filtering powered by custom DSP algorithms.
3. **Neural Decoding** — A transformer-based deep learning model decodes imagined speech patterns from brain waves.
4. **Output** — Decoded intent is rendered as text or synthesised speech in under 200 ms.

---

## Technology

- **EEG Hardware:** High-density dry-electrode cap (no gel, no surgical implants)
- **AI Model:** Transformer architecture fine-tuned on imagined speech EEG datasets
- **Edge Inference:** Runs on a dedicated edge compute module — no cloud required for core decoding
- **Connectivity:** Bluetooth 5.2 low-energy to companion app (iOS / Android / Web)
- **Accuracy:** 94.7% character accuracy on benchmark datasets (BCIC-IV, PhysioNet EEG)

---

## Why NeuroGram

| Feature | NeuroGram | Traditional AAC |
|---------|-----------|-----------------|
| Setup time | < 2 minutes | 30–60 minutes |
| Words per minute | ~40 WPM | 5–15 WPM |
| Invasiveness | Non-invasive | Often surgical |
| Internet required | No | Varies |
| Cost | ~$1,200 USD | $5,000–$15,000 |

---

## Use Cases

- ALS and locked-in syndrome patients
- Post-stroke aphasia recovery
- Children with severe cerebral palsy
- Emergency communication for first responders
- Silent communication in high-noise environments

---

## Roadmap

- **Q2 2026** — Pilot clinical trial with 30 patients across 3 hospitals
- **Q3 2026** — FDA Breakthrough Device Designation application
- **Q4 2026** — Consumer prototype launch
- **Q1 2027** — CE Mark (EU) + Health Canada approval
- **2027** — Commercial launch, starting with ALS patient programmes

---

## Pages

- [Home](https://www.nglab.tech/) — Product overview and key features
- [About](https://www.nglab.tech/about) — Team and mission
- [Contact](https://www.nglab.tech/contact) — Get in touch or request a demo

---

## Contact

- **Website:** https://www.nglab.tech/contact
- **For partnerships or clinical collaborations:** Use the contact form on the website.

---

*NeuroGram is a research-stage medical device startup. Content on this site is for informational purposes only and does not constitute medical advice.*