Video to Text Transcription Service

A Node.js application that extracts audio from video files and converts speech to text using offline transcription. This work with video where english language is used.

npm package

Features

• pass video or audio path to specific functions
• Extract audio from videos using FFmpeg
• Convert speech to text using Vosk offline speech recognition
• Local processing - no external API calls required

Prerequisites

• Node.js (v14+)
• Python 3.6+
• FFmpeg
• pip

Installation

1. Install System Dependencies

Install FFmpeg:

sudo apt install ffmpeg

Install Python dependencies:

pip install vosk

How It Works

2. Audio Extraction: extractAudio() function uses FFmpeg to extract audio from video
3. Speech Recognition: transcribeAudio() function calls Python script with Vosk to convert audio to text
4. Output: Returns transcribed text

Example

    import { extractAudio, transcribeAudio } from 'transcribe2texts';

    extractAudio("<video-path>.mp4","<path-to-save-extracted-audio>/sample.wav");

    async function fun(){

        const res=await transcribeAudio("./sample.wav");
        console.log("res:",res);
    }

    fun();

Key Technologies

• FFmpeg: Processes multimedia files, extracts audio from video
• Fluent-FFmpeg: Node.js wrapper for FFmpeg commands
• Vosk: Open-source offline speech recognition toolkit
• Child Process: Executes Python transcription script from Node.js