skip to main content

Title: Computation of Head-Related Transfer Functions Using Graphics Processing Units and a Pereptual Validation of the Computed HRTFs against Measured HRTFs
Fast generation of personalized head-related transfer functions is essential for rendering spatial audio. In this paper we propose to generate head-related transfer functions using a single graphics processing unit (GPU). We optimize the implementation of the conventional boundary element solver on a GPU. The simulation of a single frequency can be completed in seconds. A psychoacoustic experiment is conducted to study the perceptual performance of the computed HRTFs. In general, perceptual accuracy in the back is better than that in the front.
; ; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
2019 Audio Engineering Society Conference on Headphone Technology
Sponsoring Org:
National Science Foundation
More Like this
  1. Fast generation of personalized head-related transfer functions is essential for rendering spatial audio. In this paper we propose to generate head-related transfer functions using a single graphics processing unit (GPU). We optimize the implementation of the conventional boundary element solver on a GPU. The simulation of a single frequency can be completed in seconds. A psychoacoustic experiment is conducted to study the perceptual performance of the computed HRTFs. In general, perceptual accuracy in the back is better than that in the front.
  2. Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)
    The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should producemore »identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues.« less
  3. Objective: To determine if a vestibular prosthesis could improve function in subjects with severe vestibular damage and could be used it as a scientific tool to investigate central vestibular processing. Background: Damage to the vestibular labyrinth is common and usually permanent. We therefore developed and tested a vestibular implant (VI) that is designed to mimic the information normally provided by the vestibular labyrinth to determine if we can reduce vestibular-mediated deficits and study temporal integration of sensory cues in the brain. Design/Methods: Monkeys had electrodes implanted in the semicircular canals of one ear and then severe bilateral vestibular damage was induced with aminoglycosides. Eye movements, perception, and balance were tested before and after vestibular damage and with the VI activated, which supplied head motion information to the brain via electrical stimulation delivered by the implanted electrodes. Humans also had electrode implantation (done in conjunction with a cochlear implant, CI) and they were tested on a temporal binding psychophysical task Results: Stimulation provided by VI in vestibulopathic monkeys improved their balance, perception of spatial orientation, and eye movement responses. Timing experiments in humans using CI and VI stimuli showed that unlike past experiments that used motion to generate the vestibular signal,more »CI and VI signals were received by the cerebral cortex with the same latency and were perceived as simultaneous, but this timing perception was highly sensitive to adaption. Conclusions: VI improves oculomotor, postural, and perceptual behavior in vestibulopathic monkeys and could prove to be an effective way to improve these functions in patients with permanent labyrinthine damage. Timing experiments show that when novel stimuli are used, the brain synthesizes them in accordance with their arrival at the cortex, but that experience can rapidly recalibrate this timing relationship, which may be why normal stimuli that are experienced habitually lack this characteristic.« less
  4. GPU accelerators have become common on today’s leadership-class computing platforms. Effective exploitation of the additional parallelism offered by GPUs is fraught with challenges. A key performance challenge faced by developers is how to limit the time consumed by synchronizations between the CPU and GPU. We introduce the extended feed-forward measurement (FFM) performance tool that provides an automated detection of synchronization problems, identifies if the synchronization problem is a component of a larger construct that exhibits a problem beyond an individual synchronization operation, identifies remedies that can correct the issue, and in some cases automatically applies remedies to problems exhibited by larger constructs. The extended FFM performance tool identifies three causes of unnecessary synchronizations: a problem caused by a single operation, a problem caused by memory management issues, and a problem caused by a memory transfer. The extended FFM model prescribes remedies for each construct and can automatically apply remedies for memory management and memory transfer cause problems. We created an implementation of the extended FFM performance tool and employed it to identify and automatically correct problems in three real-world scientific applications, resulting in an automatically obtained reduction in execution time between 9% and 43%.
  5. The many-body correlation function is a fundamental computation kernel in modern physics computing applications, e.g., Hadron Contractions in Lattice quantum chromodynamics (QCD). This kernel is both computation and memory intensive, involving a series of tensor contractions, and thus usually runs on accelerators like GPUs. Existing optimizations on many-body correlation mainly focus on individual tensor contractions (e.g., cuBLAS libraries and others). In contrast, this work discovers a new optimization dimension for many-body correlation by exploring the optimization opportunities among tensor contractions. More specifically, it targets general GPU architectures (both NVIDIA and AMD) and optimizes many-body correlation’s memory management by exploiting a set of memory allocation and communication redundancy elimination opportunities: first, GPU memory allocation redundancy : the intermediate output frequently occurs as input in the subsequent calculations; second, CPU-GPU communication redundancy : although all tensors are allocated on both CPU and GPU, many of them are used (and reused) on the GPU side only, and thus, many CPU/GPU communications (like that in existing Unified Memory designs) are unnecessary; third, GPU oversubscription: limited GPU memory size causes oversubscription issues, and existing memory management usually results in near-reuse data eviction, thus incurring extra CPU/GPU memory communications. Targeting these memory optimization opportunities, this articlemore »proposes MemHC, an optimized systematic GPU memory management framework that aims to accelerate the calculation of many-body correlation functions utilizing a series of new memory reduction designs. These designs involve optimizations for GPU memory allocation, CPU/GPU memory movement, and GPU memory oversubscription, respectively. More specifically, first, MemHC employs duplication-aware management and lazy release of GPU memories to corresponding host managing for better data reusability. Second, it implements data reorganization and on-demand synchronization to eliminate redundant (or unnecessary) data transfer. Third, MemHC exploits an optimized Least Recently Used (LRU) eviction policy called Pre-Protected LRU to reduce evictions and leverage memory hits. Additionally, MemHC is portable for various platforms including NVIDIA GPUs and AMD GPUs. The evaluation demonstrates that MemHC outperforms unified memory management by \( 2.18\times \) to \( 10.73\times \) . The proposed Pre-Protected LRU policy outperforms the original LRU policy by up to \( 1.36\times \) improvement. 1« less