The Advanced Linux Sound Architecture is composed of several parts. The first is a fully modularized sound driver which supports module autoloading, devfs, isapnp autoconfiguration, and gives complete access to analog audio, digital audio, control, mixer, synthesizer, DSP, MIDI, and timer components of audio hardware. It also includes a fully-featured kernel-level sequencer, a full compatibility layer for OSS/Free applications, an object-oriented C library which covers and enhances the ALSA kernel driver functionality for applications (client/server, plugins, PCM sharing/multiplexing, PCM metering, etc.), an interactive configuration program for the driver, and some simple utilities for basic management.
Soundtank is a program that hosts musical instruments for you to play; each instrument can be controlled via MIDI. You can create custom instruments and save them to files, and can also load instruments that other people create. It relies on LADSA plugins, as it has no built-in sound synthesis algorithms, but hosts them in "realtime objects" which embody the structure of the audio signal flow. RTObjects can be controlled in a completely customizable fashion using MIDI events sent through the ALSA sequencer interface. Soundtank outputs audio through ALSA or JACK.
Beast is a powerful music composition and modular synthesis application. It offers multiple input methods like multitrack, piano roll, and pattern editing and supports skins. On the technical side, it has a wide range of abilities like sequencing, unlimited undo/redo history, real-time synthesis with multiprocessor support, full duplex 32-bit audio rendering, precise timing down to sample granularity, on demand loading of partial wave files, on the fly decoding of various sample formats, aliasing free oscillators, and full Scheme scripting support.
Twibright Syntring is a purely algorithmic string synthesizer with ASCII tab input and wav output. It doesn't use any samples, yet sounds quite realistic. Timbre, volume, and timing are adjustable in the tab file. It can be used for learning or practicing guitar or bass playing, for editing and synthesizing one's own tunes, or producing MIDI sound fonts. It is very lightweight and portable, and doesn't require any libraries.
Algoscore is a graphical environment for algorithmic composition, where music is constructed directly in an interactive graphical score. The result is output to MIDI and CSound. Graphical objects are placed in a timeline and connected together. Some objects are fully dependent on user data, while some are generative and reacts on input from other objects. It is highly customizable and extensible with the Nasal scripting language. It has a non-realtime perspective where the composer can relate freely to time and construct the composition outside of time. This makes AlgoScore more like a traditional sequencer in this regard, but a graphical algorithmic sequencer with powerful and flexible scripting capabilities.
The Freeverb3 library includes high quality freeverb, nreverb, impulse response reverb, FIR filter constructor, band splitter, and hardknee/softknee compressor. Its oversampling feature provides high quality audio processing. The compressor provides mastering quality audio processing. This library includes XMMS plugins, audacious plugins, BMP beep media player plugins, and sample programs, which enable you to test the audio effects easily. Freeverb3_vst is a VST plugin version of the freeverb3 library.
Sfront compiles MPEG 4 Structured Audio (MP4-SA) programs into efficient C programs that generate audio when executed. MP4-SA is a standard for normative algorithmic sound, which combines an audio signal processing language (SAOL) with score languages (MIDI and SASL). Under Mac OS X and Linux, sfront supports real-time, low-latency audio input/output and MIDI input. Under Mac OS X, sfront supports the creation of AudioUnit plug-ins. The documentation includes a book about SAOL programming.