Performance analysis and optimization of automatic speech recognition
View/Open
Cita com:
hdl:2117/128336
Document typeArticle
Defense date2018-10-01
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial
property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public
communication or transformation of this work are prohibited without permission of the copyright holder
Abstract
Fast and accurate Automatic Speech Recognition (ASR) is emerging as a key application for mobile devices. Delivering ASR on such devices is challenging due to the compute-intensive nature of the problem and the power constraints of embedded systems. In this paper, we provide a performance and energy characterization of Pocketsphinx, a popular toolset for ASR that targets mobile devices. We identify the computation of the Gaussian Mixture Model (GMM) as the main bottleneck, consuming more than 80 percent of the execution time. The CPI stack analysis shows that branches and main memory accesses are the main performance limiting factors for GMM computation. We propose several software-level optimizations driven by the power/performance analysis. Unlike previous proposals that trade accuracy for performance by reducing the number of Gaussians evaluated, we maintain accuracy and improve performance by effectively using the underlying CPU microarchitecture. First, we use a refactored implementation of the innermost loop of the GMM evaluation code to ameliorate the impact of branches. Second, we exploit the vector unit available on most modern CPUs to boost GMM computation, introducing a novel memory layout for storing the means and variances of the Gaussians in order to maximize the effectiveness of vectorization. Third, we compute the Gaussians for multiple frames in parallel, so means and variances can be fetched once in the on-chip caches and reused across multiple frames, significantly reducing memory bandwidth usage. We evaluate our optimizations using both hardware counters on real CPUs and simulations. Our experimental results show that the proposed optimizations provide 2.68x speedup over the baseline Pocketsphinx decoder on a high-end Intel Skylake CPU, while achieving 61 percent energy savings. On a modern ARM Cortex-A57 mobile processor our techniques improve performance by 1.85x, while providing 59 percent energy savings without any loss in the accuracy of the ASR system.
Description
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
CitationTabani, H. [et al.]. Performance analysis and optimization of automatic speech recognition. "IEEE Transactions on Multi-Scale Computing Systems", 1 Octubre 2018, vol. 4, núm. 4, p. 847-860.
ISSN2332-7766
Publisher versionhttps://ieeexplore.ieee.org/document/8010340
Files | Description | Size | Format | View |
---|---|---|---|---|
TMCS2018.pdf | 2,078Mb | View/Open |