The purpose of this project is to test and flesh out a feature (speech detection) for another project.

Simple Speech Detection

A Vue.js application for simple speech detection and analysis using Web Audio API, AudioWorklet, Meyda, and Chart.js. This project demonstrates real-time audio feature extraction and visualization with capabilities for recording and analyzing speech segments.

Features

• Real-time audio streaming and visualization using WaveSurfer.js (not working)
• Audio feature extraction using Meyda
• Speech detection with customizable silence threshold and cooldown
• Recording and playback of detected speech segments
• Visualization of audio features (RMS, ZCR, Spectral Centroid, Spectral Rolloff, MFCC) using Chart.js

Getting Started

Prerequisites

• Node.js (>= 14.x)
• npm

Installation

1. Clone the repository:

   git clone https://github.com/your-username/simple-speech-detection.git
   cd simple-speech-detection
   

2. Install dependencies:

   npm install
   

Running the application

1. Start the development server:

   npm run dev
   

2. Open your browser and navigate to http://localhost:8080 (or the port specified in your console).

Usage

Click anywhere on the page to initialize audio. The application will start listening to the microphone input. Detected speech segments will be recorded and displayed in a list with playback controls. Audio features of each segment will be visualized in a chart.

Customization

Adjust silenceThreshold and silenceTimeoutMs in SpeechDetection.vue to change speech detection sensitivity. Modify the preSpeechBufferSize to include more or less audio before detected speech segments.

Contributing

Although contributions are welcome, do note that this is a throwaway project and as such won't likely be maintained to any reasonable degree.

Acknowledgements

• Vue.js
• WaveSurfer.js
• Meyda
• Chart.js