{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,9,2]],"date-time":"2024-09-02T05:07:11Z","timestamp":1725253631464},"reference-count":53,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61774038","62074035"],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"China Major S&T Project","award":["2018ZX01031-101"]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE J. Solid-State Circuits"],"published-print":{"date-parts":[[2021,1]]},"DOI":"10.1109\/jssc.2020.3029097","type":"journal-article","created":{"date-parts":[[2020,10,21]],"date-time":"2020-10-21T17:39:46Z","timestamp":1603301986000},"page":"151-164","source":"Crossref","is-referenced-by-count":47,"title":["A 510-nW Wake-Up Keyword-Spotting Chip Using Serial-FFT-Based MFCC and Binarized Depthwise Separable CNN in 28-nm CMOS"],"prefix":"10.1109","volume":"56","author":[{"ORCID":"http:\/\/orcid.org\/0000-0001-5520-1326","authenticated-orcid":false,"given":"Weiwei","family":"Shan","sequence":"first","affiliation":[]},{"ORCID":"http:\/\/orcid.org\/0000-0003-1141-4818","authenticated-orcid":false,"given":"Minhao","family":"Yang","sequence":"additional","affiliation":[]},{"given":"Tao","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Yicheng","family":"Lu","sequence":"additional","affiliation":[]},{"ORCID":"http:\/\/orcid.org\/0000-0001-9794-8049","authenticated-orcid":false,"given":"Hao","family":"Cai","sequence":"additional","affiliation":[]},{"given":"Lixuan","family":"Zhu","sequence":"additional","affiliation":[]},{"ORCID":"http:\/\/orcid.org\/0000-0001-7232-7852","authenticated-orcid":false,"given":"Jiaming","family":"Xu","sequence":"additional","affiliation":[]},{"ORCID":"http:\/\/orcid.org\/0000-0001-8512-9783","authenticated-orcid":false,"given":"Chengjun","family":"Wu","sequence":"additional","affiliation":[]},{"given":"Longxing","family":"Shi","sequence":"additional","affiliation":[]},{"ORCID":"http:\/\/orcid.org\/0000-0002-8379-0321","authenticated-orcid":false,"given":"Jun","family":"Yang","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/2890498"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2712626"},{"key":"ref33","first-page":"1","article-title":"Research on pipeline R22SDF FFT","author":"li","year":"2009","journal-title":"Proc Int Radar Conf"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/78.752621"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2019.2894360"},{"key":"ref30","article-title":"Voice recognition algorithms using Mel frequency cepstral coefficient (MFCC) and dynamic time warping (DTW) techniques","author":"muda","year":"2010","journal-title":"arXiv 1003 4083"},{"key":"ref37","first-page":"230","article-title":"A 510 nW 0.41 V low-memory low-computation keyword-spotting chip using serial FFT-based MFCC and binarized depthwise separable convolutional neural network in 28 nm CMOS","author":"shan","year":"2020","journal-title":"IEEE Int Solid-State Circuits Conf (ISSCC) Dig Tech Papers"},{"key":"ref36","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","volume":"16","author":"abadi","year":"2016","journal-title":"Proc OSDI"},{"key":"ref35","author":"warden","year":"2017","journal-title":"Speech Commands A public dataset for single-word speech recognition"},{"key":"ref34","first-page":"448","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","author":"ioffe","year":"2015","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref28","first-page":"232","article-title":"A 65 nm 24.7 J\/frame 12.3 mW activation-similarity-aware convolutional neural network video processor using hybrid precision, inter-frame data reuse and mixed-bit-width difference-frame data codec","author":"yuan","year":"2020","journal-title":"IEEE Int Solid-State Circuits Conf (ISSCC) Dig Tech Papers"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2017.54"},{"key":"ref29","first-page":"1","article-title":"Feature extraction methods LPC, PLP and MFCC in speech recognition","volume":"1","author":"dave","year":"2013","journal-title":"International Journal of Advanced Research and Engineering Technology"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/SiPS.2015.7345026"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854370"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2752838"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2896880"},{"key":"ref21","article-title":"Deep speech 2: End-to-end speech recognition in English and Mandarin","author":"amodei","year":"2015","journal-title":"arXiv 1512 02595"},{"key":"ref24","article-title":"BinaryConnect: Training deep neural networks with binary weights during propagations","author":"courbariaux","year":"2015","journal-title":"arXiv 1511 00363"},{"key":"ref23","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"arXiv 1704 04861"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2616357"},{"key":"ref25","article-title":"Binarized neural networks: Training deep neural networks with weights and activations constrained to +1 or ?1","author":"courbariaux","year":"2016","journal-title":"arXiv 1602 02830 [cs]"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2005.10.010"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2018.2839347"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462688"},{"key":"ref52","first-page":"4087","article-title":"Small-footprint keyword spotting using deep neural networks","author":"chen","year":"2015","journal-title":"Proc IEEE Int Conf Acoust Speech Signal Process (ICASSP)"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9004005"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1989.266505"},{"key":"ref40","first-page":"766","article-title":"A new approach to pipeline FFT processor","author":"he","year":"1996","journal-title":"Proc IEEE Symp Parallel Distrib Process"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1990.115555"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1991.150338"},{"key":"ref14","first-page":"213","article-title":"Iterative posterior-based keyword spotting without filler models","author":"silaghi","year":"1999","journal-title":"Proc IEEE Autom Speech Recogn Understand (ASRU) Workshop"},{"key":"ref15","first-page":"1118","article-title":"Spotting subsequences matching an HMM using the average observation probability criteria with application to keyword spotting","author":"silaghi","year":"2005","journal-title":"Proc Nat Conf Artif Intell (AAAI)"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2778702"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2018.2869150"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2018.8310263"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2767705"},{"key":"ref4","article-title":"Hello edge: Keyword spotting on microcontrollers","author":"zhang","year":"2017","journal-title":"arXiv 1711 07128"},{"key":"ref3","article-title":"Small-footprint keyword spotting using deep neural network and connectionist temporal classifier","author":"wang","year":"2017","journal-title":"arXiv 1709 03665"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/VLSIC.2018.8502309"},{"key":"ref5","first-page":"250","article-title":"A 288 $\\mu\\text{W}$\n programmable deep-learning processor with 270 KB on-chip weight storage using non-uniform memory hierarchy for mobile intelligence","author":"bang","year":"2017","journal-title":"IEEE Int Solid-State Circuits Conf (ISSCC) Dig Tech Papers"},{"key":"ref8","first-page":"52c","article-title":"18 $\\mu\\text{W}$\n SoC for near-microphone keyword spotting and speaker verification","author":"giraldo","year":"2019","journal-title":"Proc IEEE Symp VLSI Circuits (VLSI-Circuits)"},{"key":"ref7","first-page":"166","article-title":"Laika: A 5 $\\mu\\text{W}$\n programmable LSTM accelerator for always-on keyword spotting in 65nm CMOS","author":"giraldo","year":"2018","journal-title":"Proc Eur Solid-State Circuits Conf (ESSCIRC)"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/29.45545"},{"key":"ref9","first-page":"1478","article-title":"Convolutional neural networks for small-footprint keyword spotting","author":"sainath","year":"2015","journal-title":"Proc Annu Conf Int Speech Commun Assoc (Interspeech)"},{"key":"ref46","article-title":"Small-footprint keyword spotting on raw audio data with sinc-convolutions","author":"mittermaier","year":"2019","journal-title":"arXiv 1911 02086"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639585"},{"key":"ref48","year":"0","journal-title":"Common Voice An Open Source Multi-Language Dataset of Voices That Anyone Can Use to Train Speech-Enabled Applications"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2011.2176163"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1988.197015"},{"key":"ref44","first-page":"1","article-title":"Learning the speech front-end with raw waveform CLDNNs","author":"sainath","year":"2015","journal-title":"Proc INTERSPEECH"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2004.837945"}],"container-title":["IEEE Journal of Solid-State Circuits"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/4\/9306027\/09233931.pdf?arnumber=9233931","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T14:55:00Z","timestamp":1652194500000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9233931\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,1]]},"references-count":53,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/jssc.2020.3029097","relation":{},"ISSN":["0018-9200","1558-173X"],"issn-type":[{"value":"0018-9200","type":"print"},{"value":"1558-173X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,1]]}}}