Audicle

Turn the Web into Your Personal Podcast with Audicle. Local-First Neural Text-to-Speech for Articles & Tweets.

Overview

Audicle is a powerful Chrome Extension that transforms written content into high-fidelity audio. Unlike traditional screen readers, Audicle focuses on listening experience and speed reading.

It bridges the gap between web content and neural audio by integrating:

1. Local Inference: Use Kokoro-82M for free, private, offline synthesis.
2. Premium Voice Quality: Optional integration with ElevenLabs for premium voices.
3. RSVP Speed Reading: Consume text at 3x speed with visual guides.

Key Features

• Local-First Architecture: Your library lives on your device. Zero tracking, zero cloud dependency (unless you choose it).
• Neural Voice Engine: Support for Kokoro-82M (via Localhost API) for near-human quality with <200ms latency.
• RSVP Speed Reader: "Rapid Serial Visual Presentation" mode synchronizes audio with single-word visual updates to reduce eye strain.
• Twitter/X Integration: Automatically scrapes metadata (Author, Avatar, Timestamp) when saving tweets.
• Analytics Dashboard: Track your listening habits, daily usage, and total words consumed.[under development]
• Beautiful UI: Crafted with Tailwind CSS, framer-motion animations, and a responsive glassmorphic design.

Getting Started

Prerequisites

• Node.js (v18+)
• pnpm or npm
• Docker (Optional, for running Kokoro server)

Installation

1. Clone the Repository

   git clone https://github.com/yourusername/audicle.git
   cd audicle

2. Install Dependencies

   npm install

3. Start Development Server

   npm run dev

   This will load the extension into Chrome. Open chrome://extensions, enable Developer Mode, and click Load Unpacked pointing to the build/chrome-mv3-dev folder.

Setup: Voice Engines

Option A: Local Inference (Recommended)

Get free, unlimited, high-quality TTS by running the model locally.

1. Clone Kokoro-FastAPI

   git clone https://github.com/remsky/Kokoro-FastAPI.git
   cd Kokoro-FastAPI

2. Run with Docker (Easiest)

   docker-compose up --build

   Or run manually with Python/uvicorn.
3. Connect Audicle
   • Go to Settings in the Dashboard.
   • Enable Kokoro TTS.
   • Endpoint is usually http://localhost:8880.

Option B: ElevenLabs Cloud

1. Get an API Key from ElevenLabs.io.
2. Go to Settings -> ElevenLabs.
3. Paste your API Key.

Architecture

graph TD
    subgraph Browser Extension
        Popup[Popup UI]
        SidePanel[SidePanel Player]
        Bg[Background Service Worker]
        Content[Content Script]
        Storage[(Chrome Storage)]
    end

    subgraph Local Machine
        Kokoro[Kokoro-FastAPI Server]
        Model[Kokoro-82M ONNX]
    end

    subgraph Cloud
        EL[ElevenLabs API]
    end

    Content -->|Scrapes Text| Bg
    Bg -->|Saves| Storage
    Bg -->|Fetches Audio| Kokoro
    Bg -->|Fetches Audio| EL
    Kokoro -->|Runs| Model

Loading

Development

This project uses Plasmo, a framework for browser extensions.

• popup.tsx: The extension popup (Quick Player).
• options.tsx: The full Dashboard (Home, Library, Analytics).
• background.ts: Service worker handling audio streaming and downloads.
• contents/: Scripts for parsing article text.

Open Source software.

Credits

• Model: Kokoro-82M
• Kokoro FastAPI: https://github.com/remsky/Kokoro-FastAPI
• Framework: Plasmo