Skip to content

Processing Modules

The Processing Modules encompass a variety of signal processing components used by high-level models to perform simulations and audio rendering tasks. These modules handle essential operations such as convolution, filtering, and encoding, acting as the core signal processing units for the BRT Library's functionality.

Modular Design

To promote flexibility and ease of development, the implementation of these modules typically separates signal processing from the connectivity required within the library. This design choice facilitates both the development of new modules and their reuse in other contexts. Each processing module is generally implemented with two distinct classes or files 1:

  • A signal processing class, which is independent of the BRT Library’s architecture but may rely on common components such as type definitions and vector classes. E.g. MySignalProcessing.hpp
  • A processor class, which extends the signal processing class by integrating connectivity features required for interaction with other library modules. E.g. MySignalProcessingProcessor.hpp

For instance, in the case of the HRTF Convolver, the class HRTFConvolver.hpp implements the convolution logic without depending on connections to any other BRT module. On the other hand, the class HRTFConvolverProcessor.hpp integrates this functionality within the library while adding the necessary mechanisms for connecting to other modules.

Reusability and Integration

This modular design not only streamlines development within the BRT Library but also enhances the reusability of signal processing components. By decoupling core signal processing functions from library-specific connectivity, these modules can be easily integrated into external systems or reused in other audio processing contexts, making them highly versatile and portable.

List of signal processor modules

  • Ambisonic Domain Convolver & Processor: The module performs ambisonic convolution for the N channels of each of the M different input sources, resulting in a monaural signal.
  • Ambisonic Encoder: Encodes sound sources into an ambisonic format for spatial audio processing and rendering.
  • Bilateral Ambisonic Encode & Processor: Process an audio source by applying delay and spatial expansion, simulating near-field effects through binaural filtering, and finally encoding the signals into left and right Ambisonic channels for spatial audio rendering.
  • Binaural Filter: Implement binaural filtering from second-order stages.
  • Directivity TF Convolver: Process audio signals by applying a source's directional transfer function (Directivity TF) to simulate realistic sound propagation.
  • Distance Attenuator: Implements distance-based attenuation to simulate the natural decrease in sound intensity over distance.
  • Uniform Partitioned Convolution: Implements efficient frequency-domain convolution using uniformly partitioned overlap-save methods, optimized for long impulse responses.
  • HRTF Convolver & Processor: Manage the convolution of the input signals, using the Uniform Partitioned Convolution.
  • Near Field Effect Processor: Integrate connectivity features to manage binaural filter processing

  1. There are processing modules that are used internally by other modules or signal processors. In this type of module, the class that adds connectivity functionality has not been implemented. ↩