OwnAudio is a platform-independent C# audio library that provides a high-level API for audio playback, recording, and processing. By default, it uses Miniaudio for audio I/O. If FFmpeg or PortAudio is installed, it automatically uses Portaudio and FFmpeg. This way, it can work with MiniAudio without any external dependencies. The implementation of MiniAudio also allowed the API to be used on mobile platforms. It is possible to manipulate audio data in real time (pitch change, tempo change, and various real-time effects). The API is able to detect musical chords from audio and create a timed list of chords. A feature has been built in to help game developers manage sound effects.
Detecting musical chords from audio data Real-time or offline chord detection from musical notes
New feature: SourceSpark, a useful resource for game developers, is ready. See description below!
Check out the sample application OwnAudioSharpDemo that demonstrates the capabilities of the OwnAudioSharp audio library through an Avalonia MVVM application using ReactiveUI. MainWindowViewModel.cs contains the core logic for audio processing, playback, effects application, and UI control.
Documentation
🎵 NEW: Audio Matchering in OwnAudioSharp!
Professional mastering automatically - single line of code!
The table below summarizes the supported operating systems, the APIs used, and their testing status.
System
APIs
Status
Windows
PortAudio 2, MiniAudio, FFmpeg 6
Tested
Linux
PortAudio 2, MiniAudio, FFmpeg 6
Tested
macOS
PortAudio 2, MiniAudio, FFmpeg 6
Tested
Android
MiniAudio
This project is tested with BrowserStack
iOS
MiniAudio
This project is tested with BrowserStack
The library will attempt to find these dependencies in standard system locations but also supports specifying custom paths.
Installation
You can add this library to your project via NuGet (when published) or by directly referencing the project.
Required Libraries
You will find the required files in the LIBS folder in a compressed file.
Extract the package appropriate for your operating system into the folder containing the compressed file.
Depending on your operating system, you will need the following:
Here's a quick example of how to use OwnAudio to play an audio file:
using Ownaudio;
using Ownaudio.Sources;
using System;
using System.Threading.Tasks;
try
{
// Initialize OwnAudio
OwnAudio.Initialize();
// Create a source manager
var sourceManager = SourceManager.Instance;
// Add an audio file
await sourceManager.AddOutputSource("path/to/audio.mp3");
// Play the audio
sourceManager.Play();
// Wait for the audio to finish
Console.WriteLine("Press any key to stop playback...");
Console.ReadKey();
// Stop playback and clean up
sourceManager.Stop();
}
catch (Exception ex)
{
Console.WriteLine($"Audio error: {ex.Message}");
}
finally
{
OwnAudio.Free();
}
Advanced Features
Mixing Multiple Audio Sources
using Ownaudio;
using Ownaudio.Sources;
// Add multiple audio files
await sourceManager.AddOutputSource("path/to/audio1.mp3", "Track1Name");
await sourceManager.AddOutputSource("path/to/audio2.mp3", "Track2Name");
// Adjust volume for individual sources
sourceManager["Track1Name"].Volume = 0.8f; // 80% volume for first source
sourceManager["Track2Name"].Volume = 0.6f; // 60% volume for second source
// Play mixed audio
sourceManager.Play();
Audio Recording
// Add an input source
await sourceManager.AddInputSource();
// Start recording to file
sourceManager.Play("output.wav", 16); // 16-bit recording
Time Stretching and Pitch Shifting
// Change tempo without affecting pitch (value range -20 to +20)
sourceManager["Track1Name"].Tempo = 10.0; // Speed up by 10%
// Change pitch without affecting tempo (value range -6 to +6 semitones)
sourceManager["Track1Name"].Pitch = 2.0; // Raise pitch by 2 semitones
Seeking Within Audio
// Seek to a specific position
sourceManager.Seek(TimeSpan.FromSeconds(30)); // Seek to 30 seconds
Audio Processing with Built-in Effects
OwnAudio includes a comprehensive effects library:
using Ownaudio.Effects;
// Apply reverb effect
var reverb = new Reverb(0.5f, 0.3f, 0.4f, 0.7f);
sourceManager.CustomSampleProcessor = reverb;
// Apply delay effect
var delay = new Delay(500, 0.4f, 0.3f, 44100);
sourceManager.CustomSampleProcessor = delay;
// Apply compressor
var compressor = new Compressor(0.5f, 4.0f, 100f, 200f, 1.0f, 44100f);
sourceManager.CustomSampleProcessor = compressor;
// Apply equalizer
var equalizer = new Equalizer(44100);
equalizer.SetBandGain(0, 100f, 1.4f, 3.0f); // Boost bass
sourceManager.CustomSampleProcessor = equalizer;
Available Effects
Reverb: Professional quality reverb based on Freeverb algorithm
Delay: Echo effect with feedback control
Distortion: Overdrive and soft clipping
Compressor: Dynamic range compression
Equalizer: 10-band parametric EQ
Chorus: Multi-voice modulation effect
Flanger: Variable delay modulation
Phaser: All-pass filter stages for phasing effect
Rotary: Rotary speaker simulation
DynamicAmp: Adaptive volume control
Enhancer: Harmonic enhancement
Overdrive: effect with tube-like saturation
Limiter: with look-ahead and smooth gain reduction
Custom Audio Processing
You can implement custom audio processing by implementing the SampleProcessorBase class:
using Ownaudio.Processors;
public class MyAudioProcessor : SampleProcessorBase
{
public override void Process(Span<float> samples)
{
// Process audio samples
for (int i = 0; i < samples.Length; i++)
{
// Example: Simple gain adjustment
samples[i] *= 0.5f; // 50% volume
}
}
public override void Reset()
{
// Reset internal state if needed
}
}
// Apply the processor to source manager
var processor = new MyAudioProcessor();
sourceManager.CustomSampleProcessor = processor;
SourceSpark
Game-Optimized Audio Effects
The SourceSpark class is a simplified audio source designed specifically for short audio clips and sound effects, making it perfect for game development. Unlike standard sources, SourceSpark loads entire audio files into memory for instant, zero-latency playback.
Key Features for Game Development
Zero-latency playback - Audio data stored in memory for instant access
Looping support - Perfect for ambient sounds like rain, wind, or engine noise
Multiple simultaneous playback - Create multiple instances for overlapping sounds
Automatic cleanup - Non-looping sounds are automatically removed when finished
Real-time effects - Pitch, tempo, and volume can be modified during playback
Basic Usage
using Ownaudio;
using Ownaudio.Sources;
// Add a simple sound effect
var gunshot = sourceManager.AddSparkSource("sounds/gunshot.wav", looping: false, volume: 0.9f);
gunshot.Play(); // Direct playback
// Add looping ambient sound
var rainSound = sourceManager.AddSparkSource("ambient/rain.wav", looping: true, volume: 0.4f);
rainSound.Play(); // Start looping
// Start the source manager to handle all sources
sourceManager.Play();
Advanced Game Audio
// Dynamic engine sound for vehicles
var engineSound = sourceManager.AddSparkSource("car_engine.wav", looping: true, volume: 0.6f);
engineSound.Play();
// Real-time sound modification based on game state
void UpdateEngineSound(float speed) {
engineSound.Pitch = speed * 0.01; // Higher pitch = faster speed
engineSound.Volume = Math.Min(0.8f, 0.3f + speed * 0.005f);
}
// Multiple overlapping gunshots - create separate instances
var pistol1 = sourceManager.AddSparkSource("weapons/pistol.wav");
var pistol2 = sourceManager.AddSparkSource("weapons/pistol.wav");
pistol1.Play(); // First shot
pistol2.Play(); // Second shot (overlapping)
Concurrent playback - Multiple instances of same sound
Memory Guidelines:
Ideal for clips under 10 seconds
Best suited for frequently used sounds
Perfect for looping ambient audio
Excellent for rapid-fire sound effects
Lifecycle Management
// Automatic cleanup for one-shot sounds
sparkSource.StateChanged += (sender, e) => {
if (sparkSource.HasFinished && !sparkSource.IsLooping) {
// Non-looping sounds are automatically removed
sourceManager.RemoveSparkSource(sparkSource);
}
};
// Manual cleanup for persistent sounds
if (sparkSource.IsLooping) {
sparkSource.Stop();
sourceManager.RemoveSparkSource(sparkSource);
}
The memory-based approach makes SourceSpark ideal for games where audio responsiveness and reliability are crucial, trading memory usage for guaranteed performance and zero-latency audio playback.
Audio Data Read
using System.Threading.Tasks;
// Add a real-time source
var realtimeSource = sourceManager.AddRealTimeSource(1.0f, 2); // Volume 1.0, stereo
// Submit audio samples in real-time
float[] samples = new float[1024]; // Your generated audio data
realtimeSource.SubmitSamples(samples);
Live Audio Streaming and Real-time Playback
The SourceSound class enables real-time audio streaming, perfect for:
Live audio synthesis
Network audio streaming
Real-time audio effects processing
Dynamic audio generation
using Ownaudio;
using Ownaudio.Sources;
using System;
using System.Threading.Tasks;
// Create a real-time audio source
var liveSource = sourceManager.AddRealTimeSource(1.0f, 2); // Volume, channels
// Example: Generate and stream sine wave in real-time
await Task.Run(async () =>
{
int sampleRate = 44100;
int frequency = 440; // A4 note
float amplitude = 0.3f;
int samplesPerBuffer = 1024;
double phase = 0;
double phaseIncrement = 2.0 * Math.PI * frequency / sampleRate;
while (liveSource.State != SourceState.Idle)
{
float[] buffer = new float[samplesPerBuffer * 2]; // Stereo
for (int i = 0; i < samplesPerBuffer; i++)
{
float sample = (float)(Math.Sin(phase) * amplitude);
buffer[i * 2] = sample; // Left channel
buffer[i * 2 + 1] = sample; // Right channel
phase += phaseIncrement;
if (phase >= 2.0 * Math.PI)
phase -= 2.0 * Math.PI;
}
// Submit samples for real-time playback
liveSource.SubmitSamples(buffer);
// Control timing for smooth playback
await Task.Delay(10);
}
});
// Start playback
sourceManager.Play();
Network Audio Streaming Example
// Example: Receive audio data from network and play in real-time
var networkSource = sourceManager.AddRealTimeSource(1.0f, 2);
// Network audio receiver (pseudo-code)
networkClient.OnAudioDataReceived += (audioData) =>
{
// Convert received network data to float array
float[] samples = ConvertBytesToFloats(audioData);
// Submit to real-time source for immediate playback
networkSource.SubmitSamples(samples);
};
sourceManager.Play();
Custom Audio Generator
using System;
using System.Threading.Tasks;
public class AudioGenerator
{
private SourceSound _source;
private int _sampleRate;
private bool _isGenerating;
public AudioGenerator(SourceManager manager, int sampleRate = 44100)
{
_sampleRate = sampleRate;
_source = manager.AddRealTimeSource(1.0f, 2);
}
public void StartGeneration()
{
_isGenerating = true;
Task.Run(async () =>
{
while (_isGenerating)
{
float[] audioBuffer = GenerateAudio(1024);
_source.SubmitSamples(audioBuffer);
await Task.Delay(5); // Smooth streaming
}
});
}
public void StopGeneration()
{
_isGenerating = false;
}
private float[] GenerateAudio(int samples)
{
// Your custom audio generation logic here
float[] buffer = new float[samples * 2]; // Stereo
// Fill buffer with generated audio data
for (int i = 0; i < samples; i++)
{
float sample = GenerateSample(); // Your generation method
buffer[i * 2] = sample; // Left
buffer[i * 2 + 1] = sample; // Right
}
return buffer;
}
private float GenerateSample()
{
// Implement your audio generation algorithm
return 0.0f;
}
}
// Usage
var generator = new AudioGenerator(sourceManager);
generator.StartGeneration();
sourceManager.Play();
Audio Data Extraction
// Load source audio data into a byte array
byte[] audioByte = sourceManager.Sources[0].GetByteAudioData(TimeSpan.Zero);
// Load source audio data into a float array
float[] audioFloat = sourceManager.Sources[0].GetFloatAudioData(TimeSpan.Zero);
Chord Detection
Real-time or offline chord detection from musical notes.
Core Components
Detectors
BaseChordDetector - Basic chord detection with major, minor, and 7th chords
ExtendedChordDetector - Adds suspended, diminished, augmented, and add9 chords
OptimizedChordDetector - Advanced detection with ambiguity handling and alternatives
RealTimeChordDetector - Continuous analysis with stability filtering
Analysis
SongChordAnalyzer - Full song analysis with timed chord progressions
ChordAnalysis - Detailed analysis results with confidence scores and explanations
Quick Start
using Ownaudio.Utilities.OwnChordDetect.Detectors;
using Ownaudio.Utilities.OwnChordDetect.Analysis;
using Ownaudio.Utilities.Extensions;
// Basic chord detection
var detector = new BaseChordDetector(confidenceThreshold: 0.7f);
var analysis = detector.AnalyzeChord(notes);
Console.WriteLine($"Detected: {analysis.ChordName} (confidence: {analysis.Confidence})");
// Extended chord detection with alternatives
var extendedDetector = new OptimizedChordDetector();
var (chord, confidence, isAmbiguous, alternatives) = extendedDetector.DetectChordAdvanced(notes);
// Real-time analysis
var realtimeDetector = new RealTimeChordDetector(bufferSize: 5);
var (stableChord, stability) = realtimeDetector.ProcessNotes(newNotes);
// Full song analysis
var songAnalyzer = new SongChordAnalyzer(windowSize: 1.0f, hopSize: 0.5f);
var timedChords = songAnalyzer.AnalyzeSong(allSongNotes);
// Complete example with SourceManager
var sourceManager = SourceManager.Instance;
await sourceManager.AddOutputSource("music.mp3", "MusicTrack");
// Detect chords from the loaded audio
var (timedChords, detectedKey, tempo) = sourceManager.DetectChords("MusicTrack", intervalSecond: 1.0f);
Console.WriteLine($"Detected Key: {detectedKey}");
Console.WriteLine($"Detected Tempo: {tempo} BPM");
foreach (var chord in timedChords)
{
Console.WriteLine($"{chord.StartTime:F1}s-{chord.EndTime:F1}s: {chord.ChordName} ({chord.Confidence:F2})");
}
Input Format
The library expects Note objects with:
Pitch - MIDI pitch number
Amplitude - Note volume (0.0-1.0)
StartTime / EndTime - Timing in seconds
Key Features
Template Matching - Uses chromagram analysis and cosine similarity
using Ownaudio.Utilities;
using System;
using System.IO;
using System.Threading.Tasks;
// Set audio data from existing float array
waveformDisplay.SetAudioData(sourceManager.Sources[0].GetFloatAudioData(TimeSpan.Zero));
// Handle playback position changes
waveformDisplay.PlaybackPositionChanged += OnPlaybackPositionChanged;
// Load directly from audio file
waveformDisplay.LoadFromAudioFile("audio.mp3");
// Load with specific decoder preference
waveformDisplay.LoadFromAudioFile("audio.mp3", preferFFmpeg: true);
// Asynchronous loading
await waveformDisplay.LoadFromAudioFileAsync("large_audio.wav");
// Loading from stream
using var fileStream = File.OpenRead("audio.mp3");
waveformDisplay.LoadFromAudioStream(fileStream);
private void OnPlaybackPositionChanged(object sender, double position)
{
// Update the actual audio playback position
sourceManager.Seek(TimeSpan.FromSeconds(position * sourceManager.Duration.TotalSeconds));
}
Properties
Property
Type
Description
WaveformBrush
IBrush
The color of the waveform
PlaybackPositionBrush
IBrush
The color of the playback position indicator
VerticalScale
double
Vertical scaling of the waveform (1.0 = original size)
DisplayStyle
WaveformDisplayStyle
The waveform display style (MinMax, Positive, RMS)
ZoomFactor
double
Zoom factor (1.0 = full view, larger values = more detailed view)
ScrollOffset
double
Horizontal scroll position (0.0 - 1.0)
PlaybackPosition
double
Current playback position (0.0 - 1.0)
Events
Event
Parameter
Description
PlaybackPositionChanged
double
Triggered when the user changes the playback position
You can configure the audio engine with specific parameters:
using Ownaudio.Engines;
// Initialize first
OwnAudio.Initialize();
// Configure output engine options
SourceManager.OutputEngineOptions = new AudioEngineOutputOptions(
OwnAudioEngine.EngineChannels.Stereo,
44100,
0.02 // Low latency
);
// Configure input engine options
SourceManager.InputEngineOptions = new AudioEngineInputOptions(
OwnAudioEngine.EngineChannels.Mono,
44100,
0.02 // Low latency
);
// Set frames per buffer
SourceManager.EngineFramesPerBuffer = 512;
Matchering
The professional audio matchering function built into OwnAudioSharp automatically analyzes and adjusts the spectral and dynamic properties of source audio to match the characteristics of a target audio file. This technology enables achieving consistent sound across musical masters or reproducing the sound of reference tracks.
Features
Intelligent EQ matching: 10-band frequency spectrum analysis and automatic equalization
Multiband compression: 4-band dynamic processing with frequency-specific settings
Psychoacoustic weighting: A-weighting and equal loudness contour consideration
Audio matchering automatically applies the most modern psychoacoustic algorithms and safe processing techniques to ensure professional-quality results for every use case.
Acknowledgements
Special thanks to the creators of the following repositories, whose code was instrumental in the development of OwnAudio:
