ffmpeg-for-lambda

FFmpeg Configuration for AWS Lambda

This configuration creates a minimal, optimized FFmpeg binary specifically tailored for audio processing in AWS Lambda environments. The build is focused on producing a small binary size (around 3.6MB) while maintaining essential audio processing capabilities.

Quick Start

1. Build the Binary

   • Go to the "Actions" tab in your repository
   • Select "ffmpeg-arm64" workflow
   • Click "Run workflow"
   • Download the artifacts from the completed run

2. Use the Binary

   • Extract the downloaded artifacts
   • Include the FFmpeg binary in your project
   • Follow the configuration-specific instructions below

Key Features

• Base Image: Built on Amazon Linux 2023 for deploying to AWS Lambda (al2023-arm64 and al2023-x86_64)
• Minimal Size: Optimized to be as small as possible while maintaining essential functionality
• Audio-Focused: Configured primarily for audio processing tasks
• Lambda-Optimized: Built specifically for AWS Lambda's environment and constraints

Included Libraries

• libfdk_aac: High-quality AAC audio encoder/decoder
• libmp3lame: MP3 audio encoder/decoder
• libopus: Opus audio codec for excellent compression

Enabled Functionality

Filters

• aresample: Audio resampling
• atempo: Audio tempo/speed adjustment
• format: Format conversion
• silencedetect: Silence detection in audio streams
• volume: Volume adjustment

Muxers (Output Formats)

• MP3
• AAC
• WAV

Encoders

• libmp3lame (MP3)
• libfdk_aac (AAC)
• aac (native AAC)

Protocols

• file (local file access only)

Demuxers (Input Formats)

• AAC
• MP3
• MOV
• MP4
• Matroska
• WAV

Decoders

• AAC
• MP3
• Opus
• FLAC
• PCM formats
• Vorbis
• WavPack

Bitstream Filters

• aac_adtstoasc: AAC ADTS to ASC conversion
• extract_extradata: Extract extradata from stream

Disabled Features

To minimize binary size, the following components are disabled:

• Debug information
• Documentation
• FFplay
• Network protocols
• Hardware acceleration
• Most filters
• Most muxers/demuxers
• Most encoders/decoders
• Input/output devices

Use Cases

This FFmpeg build is ideal for AWS Lambda functions that need to:

• Convert between audio formats (MP3, AAC, WAV)
• Adjust audio volume or speed
• Detect silence in audio files
• Extract audio from video files
• Process basic audio transformations

Example Usage

Here's an example of how to use the FFmpeg binary in a Python Lambda function to extract and convert audio:

import subprocess
import logging
import os

def extract_audio_segment(input_file: str, output_file: str, start_time: float, end_time: float) -> bool:
    """
    Extract an audio segment from a media file and convert it to MP3.

    Args:
        input_file: Path to the input media file
        output_file: Path where the output MP3 will be saved
        start_time: Start time in seconds (e.g., 10.5)
        end_time: End time in seconds (e.g., 30.0)

    Returns:
        bool: True if successful, False otherwise
    """
    ffmpeg_path = '/opt/ffmpeg'  # Path in AWS Lambda

    try:
        cmd = [
            ffmpeg_path,
            '-i', input_file,
            '-ss', f'{start_time:.3f}',
            '-to', f'{end_time:.3f}',
            '-map', '0:a',  # Only select audio streams
            '-vn',  # No video processing
            '-c:a', 'libmp3lame',
            '-b:a', '192k',
            '-y',  # Overwrite output file if exists
            output_file
        ]

        logging.info(f"Executing FFmpeg command: {' '.join(cmd)}")
        subprocess.run(cmd, check=True, capture_output=True, text=True)
        return True

    except subprocess.CalledProcessError as e:
        logging.error(f"FFmpeg error: {e.stderr}")
        return False
    except Exception as e:
        logging.error(f"Unexpected error: {str(e)}")
        return False

# Example usage in a Lambda handler
def lambda_handler(event, context):
    try:
        # Example: Extract 20 seconds of audio starting at 10 seconds
        success = extract_audio_segment(
            input_file='/tmp/input.mp4',
            output_file='/tmp/output.mp3',
            start_time=10.0,
            end_time=30.0
        )

        if success:
            return {
                'statusCode': 200,
                'body': 'Audio extraction successful'
            }
        else:
            return {
                'statusCode': 500,
                'body': 'Audio extraction failed'
            }

    except Exception as e:
        return {
            'statusCode': 500,
            'body': f'Error: {str(e)}'
        }

This example demonstrates:

• Extracting a specific time segment from a media file
• Converting to MP3 format with high-quality settings
• Proper error handling and logging
• Integration with AWS Lambda
• Using the binary from the correct Lambda path

License Notice

This build includes components under GPL and other licenses. The FDK-AAC codec is covered by a non-free license. Please review the FFmpeg License for details about licensing requirements if you distribute this binary.