Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection

In this work, we propose small footprint Convolutional Recurrent Neural Network models applied to the problem of wakeword detection and augment them with scaled dot product attention. We find that false accepts compared to Convolutional Neural Network models in a 250k parameter budget can be reduced...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-11
Main Authors: Mohammad Omar Khursheed, Christin, Jose, Kumar, Rajath, Fu, Gengshen, Kulis, Brian, Cheekatmalla, Santosh Kumar
Format: Article
Language:English
Subjects:
Artificial neural networks
Budgets
Mathematical models
Network latency
Neural networks
Parameters
Recurrent neural networks
Reduction
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Mohammad Omar Khursheed
Christin, Jose
Kumar, Rajath
Fu, Gengshen
Kulis, Brian
Cheekatmalla, Santosh Kumar
description In this work, we propose small footprint Convolutional Recurrent Neural Network models applied to the problem of wakeword detection and augment them with scaled dot product attention. We find that false accepts compared to Convolutional Neural Network models in a 250k parameter budget can be reduced by 25% with a 10% reduction in parameter size by using CRNNs, and we can get up to 32% improvement at a 50k parameter budget with 75% reduction in parameter size compared to word-level Dense Neural Network models. We discuss solutions to the challenging problem of performing inference on streaming audio with CRNNs, as well as differences in start-end index errors and latency in comparison to CNN, DNN, and DNN-HMM models.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2464482620</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2464482620</sourcerecordid><originalsourceid>FETCH-proquest_journals_24644826203</originalsourceid><addsrcrecordid>eNqNjMsKwjAUBYMgWLT_EHBdiElau68WV4IPcFlCvZW2aW7NQ3_fCn6Aq4FhzpmRiAuxSXLJ-YLEznWMMZ5teZqKiJwug9Kaloh-tK3xtEDzQh18i0ZpeoY6WAuTP4J_o-0dbdDSi7eghtY86E31MPk73YGH-rtakXmjtIP4xyVZl_trcUhGi88AzlcdBjudu4rLTMqcZ5yJ_6oPI0FAOA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2464482620</pqid></control><display><type>article</type><title>Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection</title><source>Publicly Available Content Database</source><creator>Mohammad Omar Khursheed ; Christin, Jose ; Kumar, Rajath ; Fu, Gengshen ; Kulis, Brian ; Cheekatmalla, Santosh Kumar</creator><creatorcontrib>Mohammad Omar Khursheed ; Christin, Jose ; Kumar, Rajath ; Fu, Gengshen ; Kulis, Brian ; Cheekatmalla, Santosh Kumar</creatorcontrib><description>In this work, we propose small footprint Convolutional Recurrent Neural Network models applied to the problem of wakeword detection and augment them with scaled dot product attention. We find that false accepts compared to Convolutional Neural Network models in a 250k parameter budget can be reduced by 25% with a 10% reduction in parameter size by using CRNNs, and we can get up to 32% improvement at a 50k parameter budget with 75% reduction in parameter size compared to word-level Dense Neural Network models. We discuss solutions to the challenging problem of performing inference on streaming audio with CRNNs, as well as differences in start-end index errors and latency in comparison to CNN, DNN, and DNN-HMM models.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Artificial neural networks ; Budgets ; Mathematical models ; Network latency ; Neural networks ; Parameters ; Recurrent neural networks ; Reduction</subject><ispartof>arXiv.org, 2020-11</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2464482620?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Mohammad Omar Khursheed</creatorcontrib><creatorcontrib>Christin, Jose</creatorcontrib><creatorcontrib>Kumar, Rajath</creatorcontrib><creatorcontrib>Fu, Gengshen</creatorcontrib><creatorcontrib>Kulis, Brian</creatorcontrib><creatorcontrib>Cheekatmalla, Santosh Kumar</creatorcontrib><title>Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection</title><title>arXiv.org</title><description>In this work, we propose small footprint Convolutional Recurrent Neural Network models applied to the problem of wakeword detection and augment them with scaled dot product attention. We find that false accepts compared to Convolutional Neural Network models in a 250k parameter budget can be reduced by 25% with a 10% reduction in parameter size by using CRNNs, and we can get up to 32% improvement at a 50k parameter budget with 75% reduction in parameter size compared to word-level Dense Neural Network models. We discuss solutions to the challenging problem of performing inference on streaming audio with CRNNs, as well as differences in start-end index errors and latency in comparison to CNN, DNN, and DNN-HMM models.</description><subject>Artificial neural networks</subject><subject>Budgets</subject><subject>Mathematical models</subject><subject>Network latency</subject><subject>Neural networks</subject><subject>Parameters</subject><subject>Recurrent neural networks</subject><subject>Reduction</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNjMsKwjAUBYMgWLT_EHBdiElau68WV4IPcFlCvZW2aW7NQ3_fCn6Aq4FhzpmRiAuxSXLJ-YLEznWMMZ5teZqKiJwug9Kaloh-tK3xtEDzQh18i0ZpeoY6WAuTP4J_o-0dbdDSi7eghtY86E31MPk73YGH-rtakXmjtIP4xyVZl_trcUhGi88AzlcdBjudu4rLTMqcZ5yJ_6oPI0FAOA</recordid><startdate>20201125</startdate><enddate>20201125</enddate><creator>Mohammad Omar Khursheed</creator><creator>Christin, Jose</creator><creator>Kumar, Rajath</creator><creator>Fu, Gengshen</creator><creator>Kulis, Brian</creator><creator>Cheekatmalla, Santosh Kumar</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20201125</creationdate><title>Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection</title><author>Mohammad Omar Khursheed ; Christin, Jose ; Kumar, Rajath ; Fu, Gengshen ; Kulis, Brian ; Cheekatmalla, Santosh Kumar</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_24644826203</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Artificial neural networks</topic><topic>Budgets</topic><topic>Mathematical models</topic><topic>Network latency</topic><topic>Neural networks</topic><topic>Parameters</topic><topic>Recurrent neural networks</topic><topic>Reduction</topic><toplevel>online_resources</toplevel><creatorcontrib>Mohammad Omar Khursheed</creatorcontrib><creatorcontrib>Christin, Jose</creatorcontrib><creatorcontrib>Kumar, Rajath</creatorcontrib><creatorcontrib>Fu, Gengshen</creatorcontrib><creatorcontrib>Kulis, Brian</creatorcontrib><creatorcontrib>Cheekatmalla, Santosh Kumar</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection (Proquest) (PQ_SDU_P3)</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Mohammad Omar Khursheed</au><au>Christin, Jose</au><au>Kumar, Rajath</au><au>Fu, Gengshen</au><au>Kulis, Brian</au><au>Cheekatmalla, Santosh Kumar</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection</atitle><jtitle>arXiv.org</jtitle><date>2020-11-25</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>In this work, we propose small footprint Convolutional Recurrent Neural Network models applied to the problem of wakeword detection and augment them with scaled dot product attention. We find that false accepts compared to Convolutional Neural Network models in a 250k parameter budget can be reduced by 25% with a 10% reduction in parameter size by using CRNNs, and we can get up to 32% improvement at a 50k parameter budget with 75% reduction in parameter size compared to word-level Dense Neural Network models. We discuss solutions to the challenging problem of performing inference on streaming audio with CRNNs, as well as differences in start-end index errors and latency in comparison to CNN, DNN, and DNN-HMM models.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2020-11
issn 2331-8422
language eng
recordid cdi_proquest_journals_2464482620
source Publicly Available Content Database
subjects Artificial neural networks
Budgets
Mathematical models
Network latency
Neural networks
Parameters
Recurrent neural networks
Reduction
title Small Footprint Convolutional Recurrent Networks for Streaming Wakeword Detection
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T03%3A28%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Small%20Footprint%20Convolutional%20Recurrent%20Networks%20for%20Streaming%20Wakeword%20Detection&rft.jtitle=arXiv.org&rft.au=Mohammad%20Omar%20Khursheed&rft.date=2020-11-25&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2464482620%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_24644826203%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2464482620&rft_id=info:pmid/&rfr_iscdi=true