Skip to content

CLI Usage

The EasyTranscribe command-line interface provides powerful tools for speech-to-text transcription. This guide covers all available commands and options.

Command Structure

easytranscribe <command> [options]

Global Options

Version Information

easytranscribe --version
Shows the installed version of EasyTranscribe.

Help

easytranscribe --help
easytranscribe <command> --help
Display help information for the main command or specific subcommands.

Commands

live - Live Microphone Transcription

Capture and transcribe speech from your microphone in real-time.

easytranscribe live [options]

Options

--model (default: base) : Choose the Whisper model for transcription : Choices: tiny, base, small, medium, large, turbo

--verbose : Enable detailed output showing recording status and processing information

Examples

# Basic live transcription
easytranscribe live

# Use the fastest model
easytranscribe live --model tiny

# Use the most accurate model with verbose output
easytranscribe live --model large --verbose

# Quick transcription for notes
easytranscribe live --model base --verbose

Behavior

  • Automatic Start: Recording begins when you start speaking
  • Silence Detection: Stops after 3 seconds of silence
  • Minimum Duration: Records for at least 2 seconds after speech starts
  • Real-time Feedback: Shows recording status when --verbose is used

file - Audio File Transcription

Transcribe speech from audio files supporting multiple formats.

easytranscribe file <filepath> [options]

Arguments

filepath (required) : Path to the audio file to transcribe

Options

--model (default: base) : Choose the Whisper model for transcription

--verbose : Show processing information and timing

Supported Formats

  • WAV - Uncompressed audio
  • MP3 - MPEG audio
  • M4A - Apple audio format
  • FLAC - Free lossless audio codec
  • OGG - Ogg Vorbis
  • WMA - Windows Media Audio
  • And many more supported by FFmpeg

Examples

# Transcribe a WAV file
easytranscribe file recording.wav

# Transcribe with high accuracy
easytranscribe file interview.mp3 --model large

# Transcribe with verbose output
easytranscribe file meeting.m4a --model medium --verbose

# Process multiple files with a script
for file in *.wav; do
    easytranscribe file "$file" --model base
done

logs - View Transcription History

Access and analyze your transcription history with comprehensive logging tools.

easytranscribe logs [options]

Options

--date : Show logs for a specific date (format: YYYY-MM-DD) or use 'today'

--tail : Show the last N log entries

--stats : Display transcription statistics and summaries

--list-dates : List all available log dates

Examples

# Show today's transcriptions
easytranscribe logs
easytranscribe logs --date today

# Show logs for a specific date
easytranscribe logs --date 2025-07-13

# Show last 10 transcriptions
easytranscribe logs --tail 10

# Show transcription statistics
easytranscribe logs --stats

# List all available log dates
easytranscribe logs --list-dates

Log Information

Each log entry includes: - Timestamp - When the transcription occurred - Source Type - Live microphone or file - Model Used - Which Whisper model was used - Duration - Length of audio processed - Processing Time - How long transcription took - Text Content - The transcribed text - File Path - For file transcriptions

Model Selection Guide

Choose the right model based on your needs:

Model Size Speed Use Case
tiny 39 MB ⚡⚡⚡ Quick notes, testing
base 74 MB ⚡⚡ General use (default)
small 244 MB ⚡ Better accuracy needed
medium 769 MB 🐌 High accuracy important
large 1550 MB 🐌🐌 Maximum accuracy
turbo 805 MB ⚡ Best accuracy + speed balance

Model Recommendation

  • For quick notes: Use tiny or base
  • For important content: Use medium or large
  • For real-time applications: Use turbo or base
  • For batch processing: Use large for best results

Advanced Usage Patterns

Batch Processing

# Process all audio files in a directory
for file in audio_files/*.{wav,mp3,m4a}; do
    if [ -f "$file" ]; then
        echo "Processing: $file"
        easytranscribe file "$file" --model base --verbose
    fi
done

Monitoring Logs

# Check for new transcriptions periodically
watch -n 30 'easytranscribe logs --tail 5'

Creating Transcription Reports

# Generate a daily report
easytranscribe logs --date today --stats > daily_report.txt

Scripted Voice Notes

#!/bin/bash
# voice_note.sh - Quick voice note script

echo "🎤 Recording voice note..."
TEXT=$(easytranscribe live --model base)

if [ -n "$TEXT" ]; then
    TIMESTAMP=$(date +"%Y%m%d_%H%M%S")
    echo "Voice Note - $(date)" > "note_$TIMESTAMP.txt"
    echo "$TEXT" >> "note_$TIMESTAMP.txt"
    echo "📝 Saved to: note_$TIMESTAMP.txt"
else
    echo "❌ No speech detected"
fi

Error Handling

Common Exit Codes

  • 0 - Success
  • 1 - General error (file not found, invalid arguments, etc.)
  • 2 - Audio device error
  • 3 - Model loading error

Troubleshooting Commands

Check your audio setup: 

# Test microphone with verbose output
easytranscribe live --model tiny --verbose

Verify file accessibility: 

# Check if file exists and is readable
ls -la your_audio_file.wav
file your_audio_file.wav

Test different models: 

# Try the smallest model first
easytranscribe file test.wav --model tiny

Integration with Other Tools

Pipe to Text Processing

# Transcribe and count words
easytranscribe live | wc -w

# Transcribe and save to file
easytranscribe live > transcript.txt

# Transcribe and search for keywords
easytranscribe file audio.wav | grep -i "important"

JSON Output (for scripting)

# Process logs programmatically
easytranscribe logs --date today | python parse_logs.py

This CLI interface makes EasyTranscribe powerful for both interactive use and automation scripts.