EchoLock
1. Introduction: The Vision Behind EchoLock
EchoLock is a specialized, privacy-centric Windows security utility designed to bridge the gap between
convenience and workstation protection. In an era where digital privacy is paramount, EchoLock provides
a hands-free solution for securing your computer using only the power of your
Core Privacy First:
Unlike traditional voice assistants, EchoLock is built to be a "Zero-Cloud" application.
It processes audio locally on your device, ensuring your voice never leaves your machine.
Speed & Efficiency:
Guided by the slogan "lock at the speed of sound," the app is optimized for
near-instant execution.
User-Centric Security:
By allowing users to define their own unique lock phrases, the app adds a
personalized layer of security that is difficult for others to guess or trigger accidentally.
Why EchoLock?
For developers, students, and professionals—like those studying CSE in Bhopal—maintaining security when
stepping away from a desk is vital. EchoLock replaces the need for manual keystrokes (Win + L) with a
seamless, voice-activated system that remains vigilant in the background.
EchoLock is built to deliver high-speed, voice-activated desktop security while maintaining complete user privacy. Below are the core features that define how the application works and what makes it secure, efficient, and fully offline.
Secure Voice-Activated Locking
Hands-free security that allows users to instantly lock their Windows workstation using
custom-defined voice phrases without touching the keyboard.
Hardware Interop: Communicates directly with the Windows system through
native locking mechanisms, ensuring the same level of security as a manual
Win + L operation.
Privacy-First “Zero-Cloud” Model
EchoLock performs all speech recognition and command processing locally on
your system without any internet dependency.
No Data Transmission: Audio input and voice commands are never uploaded
to external servers, ensuring complete user privacy.
Dynamic Synchronization: Custom voice commands are recognized instantly
without requiring application restart.
Grammar Isolation: The engine listens only for whitelisted phrases,
reducing false triggers from background noise.
Security Threshold: Enforces a strict confidence score to prevent
unauthorized or accidental unlock attempts.
EchoLock operates through a four-stage process that prioritizes speed, accuracy, and user privacy.
1. Audio Capture & Pre-processing
Active Listening: The application uses the System.Speech.Recognition library to monitor the default
system microphone in the background.
Local Processing: Audio signals are processed entirely on the local CPU. No data is recorded, cached, or
transmitted to any external servers, fulfilling the Zero-Cloud privacy model.
2. Recognition & Confidence Analysis
Grammar Isolation: The engine does not perform general dictation. It uses Choices and GrammarBuilder to
create a strict "whitelist" of allowed phrases defined in your commands.json file.
Scoring: For every detected sound, the Speech API (SAPI) generates a confidence score between 0.0 and
1.0.
Threshold Filtering: EchoLock enforces a strict 0.80 (80%) confidence threshold. If the recognized sound
scores 0.79 or lower, the engine rejects the input to prevent accidental locks from background noise.
3. Synchronization & Persistence
Instant Sync: When you add a new command on the Dashboard, the ReloadGrammar() method instantly updates
the engine's active vocabulary without requiring an application restart.
Registry Integration: To ensure the PC is protected as soon as you log in, EchoLock adds its executable
path to the Windows Registry (HKCU\...\Run) for automated startup.
4. System Action (The Lock)
Kernel Call: Once a command is validated, the application uses P/Invoke to communicate with the Windows
kernel.
LockWorkStation: It executes the LockWorkStation() function from the user32.dll library, which instantly
locks the PC just like the Win + L keyboard shortcut.
Privacy & Security: The EchoLock Commitment
EchoLock is designed with a Privacy-First architecture, ensuring that your voice and system remain
protected without compromising personal data.
1. Zero-Cloud Privacy Model
100% Offline Processing: Unlike standard voice assistants, EchoLock requires no internet connection.
No Data Transmission: All audio-to-text conversion is performed locally on your CPU; no voice data is
ever transmitted to external servers.
Transient Audio Handling: The application processes audio in real-time within the system's RAM. It does
not record, save, or cache your voice recordings to any storage medium.
2. Advanced Recognition Security
Strict Confidence Filtering: To prevent "false positives" (accidental locks), the engine enforces a high
0.80 (80%) confidence threshold.
Grammar Isolation: The engine utilizes Choices and GrammarBuilder to create a strict whitelist. It
exclusively listens for your defined commands and ignores unrelated background conversations.
Noise Rejection: Any sound scoring 0.79 or lower is automatically discarded by the engine logic.
3. System-Level Protection
Native Windows Locking: EchoLock communicates directly with the Windows kernel via the user32.dll
library.
Kernel Call Execution: It triggers the LockWorkStation() function, providing the same level of
cryptographic security as the manual Win + L shortcut.
Startup Integrity: The app integrates with the Windows Registry (HKCU\...\Run) to ensure protection is
active from the moment the user logs in.
4. Local Data Integrity
Encapsulated Storage: Your personalized command phrases are stored locally in a commands.json file
within the application's root directory.
JSON Serialization: Data is handled using the JavaScriptSerializer class, ensuring it remains structured
and only accessible by the EchoLock service.
Clean Uninstallation: The integrated uninstaller clears all registry entries and performs a recursive
deletion of the app folder, leaving no trace on the host system.