TrueAudio Next

SDKs & Libraries

TrueAudio Next
AMD TrueAudio Next (TAN) is a software development kit for GPU accelerated audio signal processing. As part of the LiquidVR™ technology initiative aimed at enabling a fully immersive and comfortable virtual reality experience, TrueAudio Next enables full real-time dynamic physics-based audio acoustics rendering.

Multiple audio sources represented as colored spheres.
Multiple audio sources represented as colored spheres.

TAN SDK Features

  • Fast OpenCL™ parallel batch convolution engine
  • Multi-threaded, multi-queue, real-time architecture
  • Extendable open source API
  • CPU and GPU implementations for all functions
  • Sample applications for GPU acceleration of Virtual Reality audio

Benefits

The TAN SDK provides real time audio applications the ability to leverage the OpenCL™ toolchain and to co-exist with graphics.

Accelerating VR game audio on the GPU frees CPU bandwidth for game physics and AI, while enabling advanced interactive acoustics modeling on many audio streams.

Virtual Reality audio has greatly increased compute requirements:

  • Conventional gaming audio rendering is only satisfactory for flat screen games
    • Quality of existing 3D positioning, attenuation and occlusion simulation satisfies the functional needs of FPS gamers
    • No expectation or even desire for “audio immersion” as it would not match the graphics experience on flat screens
    • Artificial, studio-derived models are used, and are sufficient for needs (audiophiles aside)
    • More realistic audio nice-to-have but not essential to the experience of a flat screen game
  • Virtual Reality demands Presence (Immersion, Suspension of Disbelief)
  • Presence requires all perceptual aspects including audio to simulate consensual reality
    • Audio approximations and artificial models break presence – no matter how good the HMD is
  • People instinctively know the difference between the auditory real world and rough approximations
    • Our ability to perceive reality through audio cues helped us survive and evolve – we are all audio experts
  • We don’t need to solve the problem of 3D audio using Head-Related Transfer Functions
    • This problem has already been solved for about 20 years and runs fine on today’s CPUs
  • Although necessary, 3D audio alone does not equal audio presence
  • Physics-derived audio benefits from real-time, time-varying convolution performed on every audio source.  TAN Convolution leverages the power of the GPU to provide reliable time-varying convolution for many audio sources.
  • Most advanced audio algorithms depend on high-performance FFT or FHT (Fast Hartley Transform).  The TAN library enables these algorithms to leverage the GPU.
  • The TAN library may be used in conjunction with Radeon Rays (formerly FireRays) as foundational accelerated algorithms for a complete physics-based audio solution.

Requirements

Related

Overview Blog
Guest Blog from Lakulish Antani, VP Engineering at Impulsonic, Inc.
GPUOpen Blogs
Video

Resources