Audio Text Analyzer

A Node.js command-line tool that converts audio files (MP3/AAC) to text and performs comprehensive text analysis including summarization, keyword extraction, sentiment analysis, and more.

Project Structure

audio-text-analyzer/
├── src/
│   └── index.js          # Main application
├── input/                # Place your audio files here
├── output/               # Analysis reports & transcripts are saved here
├── .vscode/              # VSCode configuration
│   ├── launch.json       # Debug configurations
│   └── tasks.json        # Task configurations
├── package.json
├── .gitignore
└── README.md

Prerequisites

• Node.js (v14 or higher)
• Python 3.7+ (required for Whisper)
• FFmpeg (for audio processing)

Install FFmpeg on macOS:

brew install ffmpeg

Install FFmpeg on Linux (Debian/Ubuntu):

sudo apt update && sudo apt install -y ffmpeg

Setup

1. Install Node.js dependencies:

# Using npm
npm install

# Or using yarn
yarn install

2. Set up Python virtual environment and install Whisper (CLI):

# Create virtual environment at project root (expected by src/index.js)
python3 -m venv venv

# Activate virtual environment
source venv/bin/activate   # Linux/macOS

# Install Whisper CLI
pip install openai-whisper

3. For future runs, activate the virtual environment before using the tool:

source venv/bin/activate   # Linux/macOS

Usage

Direct CLI

# Convert audio to text (auto or specified language)
node src/audioToText.js input/sample.mp3 -l en

# Save report and transcript
node src/audioToText.js input/sample.mp3 -o output/analysis.txt

# Convert SRT subtitles to plain text
node src/subtitleToText.js input/sample.srt

# Using npm scripts
npm run analyze              # Analyzes input/sample.mp3
npm run analyze:output       # Saves to output/analysis.txt

Options

• -l, --language <lang>: Language code (en, it, fr, es, de, etc.) - defaults to auto-detect
• -o, --output <file>: Save full report to file; transcript is also saved separately to output/<input_basename>.txt

Progress Tracking

The tool shows real-time progress during transcription:

Starting audio analysis...

Starting transcription...

Detected language: Italian
[0%] Transcribing audio...
[25%] Transcribing audio...
[50%] Transcribing audio...
[100%] Transcribing audio...
Analyzing text...
Generating report...
Analysis complete!

VSCode Integration

1. Place your audio file in the input/ directory
2. Use Ctrl+Shift+P → "Tasks: Run Task" → Select task
3. Or use F5 to debug with the configured launch settings

Available tasks:

• Install Dependencies: Run npm install
• Run Audio Analyzer: Analyze sample file
• Run with Output: Analyze and save to output directory

Features

• Speech-to-Text: Uses OpenAI Whisper CLI for accurate transcription with real-time progress
• Language Support: Auto-detect or manually specify language (Italian, French, English, Spanish, German, etc.)
• Summarization: Extractive summary of key sentences
• Keyword Extraction: Top 10 relevant keywords
• Sentiment Analysis: Overall sentiment with scoring
• Named Entity Recognition: People, places, organizations
• Topic Modeling: Top 5 topics based on word frequency
• Reading Statistics: Word count and estimated reading time

Output Files

Terminal Output

• Shows full report including transcript, analysis, and statistics

File Output

• Transcript file: Always saved to output/<input_basename>.txt
• Report file (when using -o option): Complete analysis report at the specified path

Supported Formats

• MP3
• AAC

Supported Languages

• English (en)
• Italian (it)
• French (fr)
• Spanish (es)
• German (de)
• And many more supported by Whisper

Notes

• Ensure the Python venv is created at the project root so the Whisper CLI is available at venv/bin/whisper as expected by the app.
• If Whisper or FFmpeg are not found, verify your environment is activated and dependencies are installed.