WhisperVoice

WhisperVoice: Covert voice notes. Encrypts text and hides it via LLM-generated acrostic sentences. Murf.ai creates natural audio. Browser extension decrypts with passcode, revealing hidden message or playing decoy for unauthorized listeners. Uses LLM, Murf.ai, STT APIs

WhisperVoice: Hide Secret Messages Inside Natural Voice Notes

Project Overview

WhisperVoice is an innovative project demonstrating a multi-layered approach to secure and discreet voice communication. It tackles the challenge of hiding the very existence of a secret message within seemingly normal voice notes, making covert communication accessible and intuitive.

The Solution: Multi-API Synergy

WhisperVoice leverages the power of three distinct APIs to achieve its goal of discreet and secure voice messaging:

• Playfair Cipher (Robust Encryption): Employs this classic symmetric encryption algorithm for strong cryptographic security. The hidden message remains unreadable without the correct "keyword."

• Large Language Model (LLM) (Steganography - Hiding in Plain Sight): The innovative core of our steganography. An LLM (e.g., ChatGPT) is prompted to generate a natural-sounding sentence where the first letter of each word spells out the Playfair ciphertext. This creates an "acrostic" poem, disguised as a normal sentence.

• Murf.ai (AI Voice Synthesis): The LLM-generated "decoy message" is converted into remarkably natural-sounding human-like speech using the Murf.ai Text-to-Speech (TTS) API. This audio file, sounding like a regular voice note, is what gets transmitted.

• Speech-to-Text (STT) API (Extraction): On the receiving end, an STT API (e.g., Google Cloud Speech-to-Text, AssemblyAI) transcribes the Murf.ai audio back into text for extraction.

How It Works (End-to-End Flow)

A. Sender's Side (Encryption, Steganography & Voice Generation)

1. Secret Message & Shared Keys: Sender inputs plaintext, a pre-shared Playfair "Keyword," and a "Passcode" for decryption.

2. Playfair Encryption: Plaintext is encrypted into a "Ciphertext" using the Playfair cipher and the Keyword.

3. LLM-Powered Steganography (The "First-Letter Code"):

   • The Playfair Ciphertext (e.g., KGYPCY) is used to prompt an LLM.

   • The LLM generates a grammatically correct "Decoy Message" (e.g., "Kids gathered, yearning playfully, crunching yummies."), where the first letter of each word spells out the ciphertext.

4. Murf.ai Voice Synthesis: This Decoy Message is sent to the Murf.ai API, transforming it into a natural-sounding audio voice note.

5. Covert Transmission: The sender shares this audio file via standard messaging platforms (e.g., WhatsApp).

B. Receiver's Side (Extraction, Decryption & Message Reveal with Browser Extension)

1. Voice Note Interception: A custom browser extension detects the incoming audio message on platforms like WhatsApp.

2. Passcode Prompt: The extension prompts the user for their shared "Passcode."

3. Conditional Playback/Decryption:

   • Incorrect Passcode: The extension simply plays the original "decoy" voice note. The listener hears a normal, but semantically irrelevant, message, unaware of any secret.

   • Correct Passcode & Extraction:

     • The audio is sent to a Speech-to-Text (STT) API for transcription back into the "Decoy Message Text."

     • The steganography rule is applied: the first letter of every word from the transcribed text is extracted (e.g., KGYPCY), reconstructing the "Playfair Ciphertext."

4. Playfair Decryption: Using the pre-shared "Keyword," the Playfair Ciphertext is decrypted, revealing the "Original Secret Message."

5. Message Reveal (Optional Voice Playback): The decrypted plaintext is presented (as text in a pop-up, or optionally converted to voice again via Murf.ai and played aloud for an immersive reveal).

Innovation & Impact

• Multi-layered Security: Combines cryptographic encryption with ingenious linguistic steganography for robust stealth.

• AI for Advanced Discretion: Leverages LLMs to generate highly convincing, natural-sounding cover messages, turning ordinary voice notes into powerful covert communication channels.

• Intuitive User Experience: The browser extension streamlines the complex process, making it user-friendly. The "sounds different" effect adds practical deception.

• API-Driven Scalability: Demonstrates robust integration with multiple external APIs (LLM, Murf.ai, STT), highlighting real-world potential.

• Hackathon Feasibility: A streamlined approach that balances advanced concepts with practical, demonstrable implementation.

Potential Use Cases

• Sensitive Information Exchange: For individuals or teams requiring deniability and discretion.

• Journalism & Activism: Protecting sources and sensitive information in high-stakes environments.

• Personal Privacy: A novel and secure way to exchange private messages, adding intrigue to communication.

• WhisperVoice Chrome Extension

Covert Voice Message Encryption and Steganography

---

🚀 Installation

To install both the sender and receiver extensions:

1. Clone this repository or download the source code as a ZIP file and extract it.

2. Open your Chrome browser and navigate to chrome://extensions/.

3. Enable Developer mode in the top right corner.

4. Click Load unpacked.

5. Select the folder containing the extension files:

(This should be the folder containing manifest.json)

6. The WhisperVoice extension (sender and receiver) will now appear in your browser.

---

⚙️ Configuration

Both sender and receiver require API keys for external services.

🎯 Get API Keys

LLM API Key (e.g., Gemini)

1. Go to Google AI Studio (or your chosen LLM provider).
2. Sign in with your Google account.
3. Create a new API key.
4. Copy the key for use in the extension.

Murf.ai API Key

1. Sign up at Murf.ai.
2. Subscribe to a plan that includes API access.
3. Go to your account settings to find your API key.
4. Copy the key for use in the extension.

Speech-to-Text API Key (e.g., AssemblyAI)

1. Sign up at AssemblyAI (or your chosen STT provider).
2. Get your API key from the dashboard.
3. Copy the key for use in the extension.

---

🔧 Configure the Extension

1. Click the WhisperVoice icon in your browser toolbar.
2. In the settings interface:

• Enter your LLM (Gemini/ChatGPT) API key.
• Enter your Murf.ai API key.
• Enter your STT (AssemblyAI) API key.
• Set the Playfair Keyword (shared secret) for encryption/decryption.

3. Click Save Keys/Settings.

✅ You are now ready to send and receive covert voice messages!

---

✉️ How to Use

A. Sending a Covert Voice Message

1. Enter Your Secret Message

Type the confidential message into the sender interface.

2. Set Encryption Parameters

• Ensure your Playfair keyword is correctly configured.
• Set a passcode (required for the receiver).

3. Choose Cover Theme (Optional) Select a theme (e.g., work, college, family) to guide the LLM in generating a decoy sentence.

4. Select Voice (Optional) Pick from available Murf.ai voice options.

5. Generate

• Encrypt your message with the Playfair cipher.
• Generate a decoy sentence where the first letters encode the ciphertext.
• Convert it to audio via Murf.ai.

6. Share Download the audio file (MP3, WAV, etc.) and share via WhatsApp, email, etc.

---

B. Receiving and Decrypting a Covert Voice Message

You can start decryption in two ways:

Method 1: Floating Button

• Navigate to any website.
• Look for the blue WhisperVoice button in the bottom-right corner.
• Click to open the upload modal.

Method 2: Extension Popup

• Click the WhisperVoice icon in the toolbar.
• Use the popup for quick access.

---

🛡️ Decryption Process

1. Upload Audio

• Drag & drop or browse to select the audio file.
• Supported formats: MP3, WAV, M4A, OGG, WebM.

2. Enter Passcode

• Input the passcode shared by the sender.

3. Click Decrypt

• The extension will:
  • Transcribe the audio via the STT API.
  • Extract the hidden ciphertext.
  • Decrypt using the Playfair keyword.
  • (Optional) Convert the decrypted text back to speech.

4. View Results

• See the decrypted secret message.
• Optionally play the synthesized audio.

---

🧠 Technical Details & How It Works

• Encryption: Playfair cipher using a shared keyword.
• Steganography:
• Encoding: LLM generates a sentence where the first letter of each word encodes the ciphertext.
• Decoding: The extension extracts the first letters after transcription.
• Voice Synthesis: Murf.ai API.
• Speech-to-Text: AssemblyAI (or your chosen STT provider).

---

🎵 Supported Audio Formats

• MP3
• WAV
• M4A
• OGG
• WebM

---

🔒 Privacy & Security

• Multi-layer Protection: Encryption + steganography.
• Passcode Protection: Extra layer on the receiver side.
• Natural Cover: AI-generated sentences avoid suspicion.
• Local Processing: Encryption/decryption happens in-browser.
• Secure API Key Storage: Stored in Chrome's local storage.
• No Permanent Storage: Audio and transcriptions are discarded after processing.

---

💻 User Interface Highlights

• Floating Button: Quick access from any page.
• Upload Modal: Drag & drop files, enter passcode, see progress.
• Results Display: Decrypted message, optional audio playback.

---

🛠️ Troubleshooting

Common Issues:

• Upload Fails: Check file format, size, and connection.
• API Errors: Verify keys and API quotas.
• Decryption Fails: Ensure matching Playfair keyword and clear audio.
• No Speech Detected: Try clearer audio or another file.

Error Examples:

• Configuration Required: API keys missing.
• Invalid file format: Unsupported audio.
• No speech detected: STT could not transcribe.
• Decryption failed: Wrong passcode or keyword.

---

Key Components

• WhisperVoice Core: Encryption, steganography, API integration.
• PlayfairCipher Class: Cryptographic operations.
• UI Modules: Sender interface, upload modal, results display.
• API Modules: Gemini/Murf/STT communication.

---

⚠️ Disclaimer

This extension is for educational, personal, and legitimate privacy uses only. You are responsible for complying with applicable laws and third-party API terms of service.

---

📝 Version History

v1.0.0 (Initial Release)

• Playfair cipher encryption/decryption.
• LLM-powered first-letter steganography.
• Murf.ai TTS integration.
• AssemblyAI STT integration.
• Chrome extension sender/receiver UI.
• API key management.
• Basic error handling and troubleshooting.
• Multi-format audio support.

---