Thierry Tambe, En-Yu Yang, Glenn G. Ko, Yuji Chai, Coleman Hooper, M. Donato, P. Whatmough, Alexander M. Rush, D. Brooks, Gu-Yeon Wei
{"title":"SM6:用于精确和噪声鲁棒的基于注意力的NLP应用的16nm片上系统:第33届热芯片研讨会(2021年8月22-24日","authors":"Thierry Tambe, En-Yu Yang, Glenn G. Ko, Yuji Chai, Coleman Hooper, M. Donato, P. Whatmough, Alexander M. Rush, D. Brooks, Gu-Yeon Wei","doi":"10.1109/HCS52781.2021.9567180","DOIUrl":null,"url":null,"abstract":"In this work, we present SM6, an SoC architecture for real-time denoised speech and NLP pipelines, featuring (1) MSSE: an unsupervised probabilistic sound source separation accelerator, (2) FlexNLP: a programmable inference accelerator for attention-based seq2seq DNNs using adaptive floating-point datatypes for wide dynamic range computations, (3) a dual-core Arm Cortex A53 CPU cluster, which provides on-demand SIMD FFT processing, and operating system support. In adverse acoustic conditions, MSSE allows FlexNLP to store up to 6x smaller ASR models obviating the very inefficient strategy of scaling up the DNN model to achieve noise robustness. MSSE and FlexNLP produce efficiency ranges of 4.33-17.6 Gsamples/s/W and 2.6-7.8TFLOPs/W, respectively, with per-frame end-to-end latencies of 15-45ms.","PeriodicalId":246531,"journal":{"name":"2021 IEEE Hot Chips 33 Symposium (HCS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"SM6: A 16nm System-on-Chip for Accurate and Noise-Robust Attention-Based NLP Applications : The 33rd Hot Chips Symposium – August 22-24, 2021\",\"authors\":\"Thierry Tambe, En-Yu Yang, Glenn G. Ko, Yuji Chai, Coleman Hooper, M. Donato, P. Whatmough, Alexander M. Rush, D. Brooks, Gu-Yeon Wei\",\"doi\":\"10.1109/HCS52781.2021.9567180\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this work, we present SM6, an SoC architecture for real-time denoised speech and NLP pipelines, featuring (1) MSSE: an unsupervised probabilistic sound source separation accelerator, (2) FlexNLP: a programmable inference accelerator for attention-based seq2seq DNNs using adaptive floating-point datatypes for wide dynamic range computations, (3) a dual-core Arm Cortex A53 CPU cluster, which provides on-demand SIMD FFT processing, and operating system support. In adverse acoustic conditions, MSSE allows FlexNLP to store up to 6x smaller ASR models obviating the very inefficient strategy of scaling up the DNN model to achieve noise robustness. MSSE and FlexNLP produce efficiency ranges of 4.33-17.6 Gsamples/s/W and 2.6-7.8TFLOPs/W, respectively, with per-frame end-to-end latencies of 15-45ms.\",\"PeriodicalId\":246531,\"journal\":{\"name\":\"2021 IEEE Hot Chips 33 Symposium (HCS)\",\"volume\":\"12 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 IEEE Hot Chips 33 Symposium (HCS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/HCS52781.2021.9567180\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE Hot Chips 33 Symposium (HCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HCS52781.2021.9567180","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
SM6: A 16nm System-on-Chip for Accurate and Noise-Robust Attention-Based NLP Applications : The 33rd Hot Chips Symposium – August 22-24, 2021
In this work, we present SM6, an SoC architecture for real-time denoised speech and NLP pipelines, featuring (1) MSSE: an unsupervised probabilistic sound source separation accelerator, (2) FlexNLP: a programmable inference accelerator for attention-based seq2seq DNNs using adaptive floating-point datatypes for wide dynamic range computations, (3) a dual-core Arm Cortex A53 CPU cluster, which provides on-demand SIMD FFT processing, and operating system support. In adverse acoustic conditions, MSSE allows FlexNLP to store up to 6x smaller ASR models obviating the very inefficient strategy of scaling up the DNN model to achieve noise robustness. MSSE and FlexNLP produce efficiency ranges of 4.33-17.6 Gsamples/s/W and 2.6-7.8TFLOPs/W, respectively, with per-frame end-to-end latencies of 15-45ms.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
