This demo allows you to try out the new AKSampler instrument. Unlike the old "AKSampler", which was just a wrapper around the AUSampler Audio Unit built into macOS and iOS, the new one is built entirely from scratch.

Inside either the iOS or macOS directory, depending on which version of the app you're installing, run the command line:

> pod install

Then open the generated workspace file:

> open SamplerDemo.xcworkspace

The demo samples are the same as the ones included in the AudioKit ROMPlayer code repo on GitHub. They have been compressed using WavPack, and a .sfz metadata file has been added, to specify key/velocity mapping and loop points.

Preparing samples for any sampler instrument is a somewhat complex process. The basic steps are:

1. Acquire sample files, either by recording a hardware or software instrument yourself, or purchasing ready-made samples.
2. If necessary, edit the samples to your liking.
3. If applicable, identify loop-start and end points for each sample file. With commercial samples, this will usually have been done for you.
4. (Optional) Create a metadata file to define the mapping of MIDI note and velocity values to specific samples. Commercial samples typically include such metadata files, often in multiple formats.
5. There is a step 5 on iOS which appears in the README file in the iOS directory.

Use of a metadata file is optional. Refer to function loadAndMapCompressedSampleFiles() in Conductor.swift to see how you can load individual sample files programmatically, with the necessary key- and velocity-mapping numbers expressed as parameter values the function calls.

If you want to work with uncompressed samples in standard formats such as .wav, .aif, .caf etc., you can call AKSampler's loadAKAudioFile() method, passing in an AKAudioFile object initialized with the URL for your file.

AKSampler presently uses the open-standard SFZ format for metadata. The demo samples package includes a README with details about how the samples and .sfz metadata file were prepared, and a useful Python3 script to convert ESX24 metadata files (.esx files) to SFZ format.

Note that AKSampler's ability to parse .sfz files is very limited. It won't work on most .sfz files.

In a pinch, you can use a text editor to examine the demo .sfz files (they're just text) and modify them appropriately for your own samples.

Additional information pertaining to iOS appears in the README file in the iOS directory.