OneDialect — Unified Assistive Communication System (UACS)

Anand P S — Sun, 03 May 2026 15:32:12 +0000

A production-grade embedded system enabling communication across speech, text, Morse, and haptic signals within a single unified pipeline.

Project Links

Official Project Page: https://anandps.in/projects/unified-assistive-communication-system
GitHub Repository: https://github.com/anand-ps/unified-assistive-communication-system

Problem

Assistive communication systems are fragmented.

Most tools solve isolated problems such as speech, hearing, or vision, forcing users to:

Switch between multiple interfaces
Learn different interaction models
Depend on external assistance

The real failure is not missing features.

It is the absence of a unified system.

Solution

UACS introduces a single communication pipeline:

Speech ⇄ Text ⇄ Morse ⇄ Haptic

Morse code acts as a consistent encoding layer, enabling deterministic translation across all modalities.

System Architecture

The system is built using a dual-layer architecture:

Processing Layer (Raspberry Pi)

Speech capture
Speech-to-text conversion
Text processing
Morse encoding

Interaction Layer (ATmega328P)

Morse input detection
Timing-based decoding
Haptic feedback control
Audio feedback
Bluetooth communication

Why this split?

Ensures real-time deterministic behavior in the interaction layer
Prevents OS-level variability from affecting timing-sensitive operations
Uses a dedicated MCU to handle sensor input, Morse timing, and haptic driver control with precise timing guarantees
Offloads computational tasks (speech processing, encoding) to the SoC (Raspberry Pi), avoiding overload on the interaction layer
Improves portability by keeping the MCU-based unit as a compact HMI device, while heavy processing remains external
Enables clear separation of concerns: compute-intensive vs time-critical execution

Operational Flow

Speech → Haptic Output

Capture speech
Convert to text
Encode into Morse
Transmit via Bluetooth
Render as vibration patterns

Haptic Input → Speech

User inputs Morse via SPDT switch
Decode timing into text
Convert to speech
Output audio

Result

A full-duplex communication loop across speech and tactile interaction.

Embedded Implementation

Built around the ATmega328P for real-time control:

Interrupt-driven Morse decoding
Timing-based dot/dash classification
PWM-controlled vibration output
UART-based Bluetooth communication
Battery-optimized operation

Hardware protections

Overcharge protection
Deep discharge protection
Short-circuit protection
Thermal protection

Hardware Stack

Raspberry Pi 4 Model B
ATmega328P microcontroller
HC-05 Bluetooth module
Linear Resonant Actuator (haptic motor)
Active buzzer
SPDT switch
Li-ion battery

Software and Control Logic

The system follows an event-driven, state-based architecture:

Input capture → decode → state transition
Morse segmentation using timing windows
Serial communication handling
Output scheduling

Processing pipelines

Speech-to-text
Text-to-Morse
Morse-to-text
Text-to-speech

Cost

Approximately ₹16,700 total.

Raspberry Pi is the primary cost driver
Remaining components are optimized for affordability

Validation

Tested under real-world conditions for:

Low-latency communication
Accurate Morse encoding and decoding
Stable Bluetooth transfer
Consistent haptic and audio feedback

Future Scope

Multi-language support
Wearable form factors
Emergency communication systems
IoT-integrated assistive devices

Closing

Assistive systems fail when they optimize isolated features instead of system behavior.

UACS solves this by treating communication as a unified, deterministic pipeline where speech, text, Morse, and haptic signals operate as one cohesive system.

— Anand P S

DEV Community: Anand P S