Audio Toolbox

Design, analyze, and simulate acoustic, speech, and audio processing systems

Audio Toolbox provides signal processing, analysis, and simulation tools for audio, speech, and acoustics. It includes audio processing algorithms, psychoacoustic metric estimation, acoustic propagation modeling, data labeling and augmentation, and feature extraction. The toolbox also includes pretrained machine learning and deep learning models for transfer learning and tasks such as classification, transcription, synthesis, and source separation. 

You can stream low-latency audio to prototype algorithms and measure signals in real time. The toolbox enables you to simulate audio and acoustic systems by combining physical acoustics, audio DSP, and psychoacoustic evaluation to assess perceptual effects earlier in the design process. You can validate designs by generating VST or Audio Unit plugins, and process audio samples using AI models and hosted audio plugins. 

Streaming Acquisition and Playback

Stream low-latency audio from sound cards using standard audio drivers (such as ASIO, WASAPI, CoreAudio, and ALSA) on Windows^®, Mac^®, and Linux^® operating systems. Use the audiostreamer object to process audio in real time, measure impulse responses, and build responsive audio apps.

Real-Time Audio in MATLAB

Documentation | Examples

Waveform with two speakers, separated speaker signals, and residual signal, showing reconstruction of the original waveform.

AI for Audio, Speech, and Acoustics

Apply deep learning and machine learning to audio, speech, and acoustics. Create, label, and augment data for model training and transfer learning. Extract features and run time-frequency transformations. Transcribe and synthesize speech using cloud or local models.

Pretrained AI Models

Documentation | Examples

Audio Processing Algorithms

Generate standard waveforms, apply common audio effects, and design audio processing systems with dynamic parameter tuning and live visualization in MATLAB and Simulink.

Parametric Equalizer Design

Documentation | Examples

A Simulink model of an active noise control system based on the filtered-X LMS algorithm and configured to run natively on a real-time Speedgoat machine.

System Modeling with Simulink

Design system models using libraries of audio processing blocks for Simulink. Tune parameters and visualize system behavior using interactive controls and dynamic plots. Simulate DSP systems, analog circuits, and deep learning models. Generate code for embedded systems.

Documentation | Examples

Simscape model of an active loudspeaker including mechanical, electrical, and acoustic physical domains, with DSP preprocessing blocks.

Lumped-Parameter Acoustics

Simulate acoustic systems using lumped-parameter models in Simulink with Simscape and the Simscape Acoustic add-on. Model enclosures, tubes, and transducers, and integrate with electrical, mechanical, and DSP components for early system-level validation.

Documentation | Examples

3D model of an office space with acoustics source and receivers.

Acoustic Room Response Models

Model room acoustics with ray tracing and image-source methods. Measure acoustic and audio system responses using audio hardware with MLS and ESS sequences. Encode and decode spatial audio with HRTFs, ambisonics, and SOFA files. Model underwater acoustic channels.

Documentation | Examples

Psychoacoustics and Audio Quality

Apply sound pressure level (SPL) meters and octave and weighting filters to audio signals. Analyze recordings using standard perceptual models for loudness, roughness, and sharpness. Evaluate audio quality and speech intelligibility with metrics such as ViSQOL, STOI, STI, and SII.

Documentation | Examples

Audio Plugin Creation and Hosting

Create interfaces for tunable algorithm parameters. Generate VST and Audio Unit plugins directly from MATLAB code. Use external VST and Audio Unit plugins as regular MATLAB objects to process signal samples in MATLAB. Control plugin parameters programmatically and through MIDI controls.

Design Your First Audio Plugin Using MATLAB

Documentation | Examples

Target Embedded and Real-Time Audio Systems

With MATLAB and Simulink coder products, generate C and C++ source code from signal processing and machine learning algorithms provided as toolbox functions, objects, and blocks. Generate CUDA^® source code from select feature extraction functions. Prototype audio processing designs on Raspberry Pi^®, Android^® or iOS devices, Speedgoat^® audio machines, and ST Discovery boards.

Documentation | Examples

Product Resources:

Documentation Examples Videos Technical articles Functions Hardware support Requirements Release notes

Audio Toolbox FAQs

Audio Toolbox provides signal processing and analysis tools for audio, speech, and acoustics, including algorithms for processing signals, estimating acoustic metrics, labeling and augmenting audio data sets, and extracting audio features.

You can stream low-latency audio to and from sound cards using standard audio drivers like ASIO, WASAPI, CoreAudio, and ALSA across Windows, Mac, and Linux operating systems, processing live audio in MATLAB with milliseconds of round-trip latency.

Yes, you can generate VST plugins, AU plugins, and standalone executable plugins directly from MATLAB code without requiring manual design of user interfaces.

The toolbox includes pretrained machine learning and deep learning models that support transfer learning for tasks such as embedding extraction, sound classification, speaker verification, speech transcription and synthesis, speech source separation, and background noise reduction.

Yes, you can design system models using libraries of audio processing blocks for Simulink, tune parameters, visualize system behavior, and simulate DSP, analog circuits, and deep learning models.

Yes, Audio Toolbox offers plugin hosting so you can process MATLAB arrays using external VST and AU plugins as regular MATLAB objects, changing plugin parameters programmatically, with user interfaces, or via MIDI controls.

You can measure impulse responses using maximum-length sequences and exponential swept sines (including nonlinearities), read and write SOFA files, analyze head-related transfer functions, encode and decode ambisonic formats, and apply loudness meters and octave filters.

Audio Toolbox provides a collection of psychoacoustic and perceptual metrics for noise impact assessment, sound quality, and speech intelligibility. These include loudness measurements, along with sound pressure level (SPL) and peak/true-peak programme metering for monitoring and normalization. For perceptual assessment beyond level, Audio Toolbox supports metrics for speech intelligibility (for example, STOI, STI, and SII) and perceived audio quality (ViSQOL). These tools enable objective, perceptually relevant evaluation of audio signals for applications in sound quality analysis.

Yes, the toolbox supports acoustic physical system simulation through complementary tools. The Simscape Acoustic add-on enables lumped-parameter modeling of physical acoustic components such as ducts, enclosures, loudspeakers, and microphones using the Simscape engine. In addition, Audio Toolbox provides room-acoustics simulation using image-source and stochastic ray-tracing methods.

Yes, with MATLAB and Simulink coder products, you can generate C and C++ source code and prototype audio processing designs on platforms such as Raspberry Pi, Android or iOS devices, Speedgoat audio machines, and ST Discovery boards.