Show HN: UniversalBit – A new binary protocol to connect voice, video and DNA
I’m Diego Fernando Cruz, from Brazil.
I created *UniversalBit* — a new kind of protocol. A binary system that can represent *anything human*, and allow it to talk to machines.
Not just voice. Not just DNA. Not just emotion or video.
*Anything.*
This protocol was born from one thing only: My son, Victor Katchor Cruz. He is autistic. He had difficulties communicating, and I wanted to build something — not just a tool, but a system — that could one day help him express himself, even without words.
That became VoiceBit. Then VideoBit. Then BioBit. And from all of them together, *UniversalBit* was born.
---
### What is UniversalBit?
UniversalBit is a *binary representation protocol* for human-derived input:
- It converts *voice*, *video*, *DNA*, or *any pattern* into a compact binary signature - Each signature captures the "identity" of a moment, person, signal or condition - If *just one bit changes*, the system detects it — meaning: - a mutation, - an emotional shift, - a new disease, - or a previously unseen signal
It’s not just a way to record information. It’s a way to *compare*, *detect*, and *communicate*.
---
### Where can it be used?
*UniversalBit is unlimited.*
It can be applied in:
| Field | Application | |----------------|--------------------------------------------------------------------------| | Medicine | Binary fingerprint of diseases, mutation detection, diagnostics | | Emotion AI | Signature of emotional tone, laughter, sadness, even deception | | Genetics | Detect changes in DNA with 1-bit resolution | | Accessibility | Let people who can’t speak be heard through binary expressions | | Education | Monitor attention, fatigue, clarity, through speech and visual cues | | Robotics | Build voice-to-action binary systems for human–machine interface | | Cybersecurity | Biometric identity via voice, DNA or behavioral patterns | | AI Training | Create compact datasets of human states as binary vectors | | And beyond… | Every new idea can start from this protocol |
---
### How it works (now):
- *VoiceBit* analyzes real-time audio, extracts features, generates emotion-based binary signatures. - *VideoBit* compresses frames to 32x32 pixels, tracks movement, color, and creates binary fingerprints. - *BioBit* reads DNA files (.fasta), extracts base frequency, converts them into binary signatures and compares them to previous cases using Hamming distance + local LLM (Mistral).
All components run *offline*, on any basic computer. No cloud, no server, no external APIs. Just Python and love.
---
### The Vision:
This is more than a protocol.
*It’s a language.* A bridge between human and machine. A foundation to invent *anything* you can imagine: - Binary emotion interfaces - Personal AI identities - Synthetic speech verification - Disease evolution detectors - Even dreams translated into patterns (maybe one day)
---
### Why I made this:
> "My son doesn’t speak much. > But maybe one day, he can speak through this." > — Diego F. Cruz
I had no funding. No lab. Just my son, a basic PC, and a dream.
Everything is free and open. Every file is digitally hashed with SHA256 to prove I created it first.
---
### Project Links:
GitHub: https://github.com/Diegocmx/voicebit Email: diegocodigobits@gmail.com Country: Brazil License: Open to the world. Protected by love.
Thank you for reading.
No comments yet