EchoMind 🧠

EchoMind is a Chrome extension that allows you to record audio from any webpage, transcribe it using Groq's blazing-fast implementation of the Whisper model, and then process the transcribed text using powerful LLMs hosted by Cerebras or Google. It's your personal AI assistant for the web!

Features

• Tab Audio Recording: Capture audio directly from your active browser tab. No need for external recording software.
• Fast Transcription: Leverage Groq's incredibly fast implementation of OpenAI's Whisper-large-v3-turbo model for accurate and speedy speech-to-text conversion.
• Multilingual Support: Transcribe audio in multiple languages, including Chinese (zh), English (en), Spanish (es), French (fr), German (de), Japanese (ja), Korean (ko), and Russian (ru). Automatic language detection is also supported.
• AI-Powered Processing: Process the transcribed text using:
  • Meta's llama-3.3-70b (hosted by Cerebras)
  • Google's gemini-2.0-flash-exp
• AI-Processing Options:
  • Summarization: Get concise summaries of lengthy audio content.
  • Question Answering: Ask questions about the audio and receive AI-generated answers.
  • Custom Prompts: Craft your own custom prompts to tailor the AI's processing to your specific needs.
• User-Friendly Interface: A clean and intuitive popup interface makes recording, transcribing, and processing simple.
• Minimize Mode: Minimize the popup to a discreet microphone icon while recording, keeping your workspace uncluttered.

Installation

There are two ways to install EchoMind:

1. From GitHub Releases (Recommended for most users):

• Go to the Releases page for this repository.
• Download the latest EchoMind-vX.X.zip file (where X.X is the version number).
• Open Chrome and navigate to chrome://extensions.
• Enable "Developer mode" in the top right corner (if it's not already enabled).
• Drag and drop the downloaded EchoMind-vX.X.zip file onto the chrome://extensions page. Chrome will install the extension.
• If Developer mode is off, you may need to unpack the zip file first, and then "Load Unpacked".

2. From Source (For developers or advanced users):

• Clone the Repository:

  git clone https://github.com/Franklyc/EchoMind.git

• Load the Extension in Chrome:

  • Open Chrome and navigate to chrome://extensions.
  • Enable "Developer mode" in the top right corner.
  • Click "Load unpacked".
  • Select the EchoMind directory you just cloned.

After either installation method:

3. Obtain API Keys:

   • Groq: Groq Cloud
   • Cerebras: Cerebras Cloud
   • Google: Google AI Studio

4. Configure API Keys:

   • Click the EchoMind extension icon in your Chrome toolbar.
   • Click the "Settings" gear icon.
   • Enter your Groq, Cerebras, and Google API keys in the respective fields.
   • Click "Save Keys".

Usage

1. Start Recording:

   • Navigate to the webpage with the audio you want to capture.
   • Click the EchoMind extension icon.
   • Select your desired transcription language (or leave it on "Auto" for automatic detection).
   • Click the "Start" button. The button will change to indicate recording is in progress.

2. Stop Recording:

   • Click the "Stop" button. EchoMind will automatically send the recorded audio to Groq for transcription.

3. View Transcript:

   • The transcribed text will appear in the "Transcript" section.

4. Process with AI:

   • Choose an AI Model: Select from the dropdown.
   • Choose a "Processing Option":
     • Summarize: Generates a summary of the transcript.
     • Answer Questions: Prepares the AI to answer questions based on the transcript. You'll likely want to follow this up with a custom prompt.
     • Custom Prompt: Enter your own instructions for the AI.
   • (Optional) If you selected "Custom Prompt", enter your prompt in the "Custom Prompt" textarea.
   • Click the "Process" button. The AI's output will appear in the "AI Output" section.

5. Minimize/Restore:

   • Click the minimize button on top right corner.
   • Click the microphone icon to restore the extension popup.

Permissions Explained

• activeTab: Required to access the currently active tab for audio capture.
• storage: Used to store your API keys, selected AI model, and the transcribed text locally.
• tabCapture: Enables capturing audio from the current tab.
• host_permissions:
  • *://api.groq.com/*: Allows the extension to communicate with the Groq API for transcription.
  • *://api.cerebras.ai/*: Allows the extension to communicate with the Cerebras API for text processing.
  • *://generativelanguage.googleapis.com/*: Allows the extension to communicate with the Google Generative Language API for text processing.

License

This project is licensed under the MIT License - see the LICENSE file for details. Copyright (c) 2025 Franklyc.

Contributing

Contributions are welcome! If you find a bug or have a feature request, please open an issue. If you'd like to contribute code, please fork the repository and submit a pull request.

Acknowledgements

• This extension utilizes the powerful and fast Whisper model (developed by OpenAI) and served by Groq.
• AI processing is powered by Meta's Llama 3 model, hosted by Cerebras, and Google's Gemini model.
• Font Awesome icons are used for a visually appealing UI.