{"id":333239,"date":"2017-02-14T05:02:10","date_gmt":"2017-02-13T21:02:10","guid":{"rendered":"http:\/\/techcrunch.com\/?p=1452013"},"modified":"2017-02-14T05:02:10","modified_gmt":"2017-02-13T21:02:10","slug":"mit-develops-a-speech-recognition-chip-that-uses-a-fraction-of-the-power-of-existing-technologies","status":"publish","type":"post","link":"https:\/\/people.utm.my\/asmawisham\/mit-develops-a-speech-recognition-chip-that-uses-a-fraction-of-the-power-of-existing-technologies\/","title":{"rendered":"MIT develops a speech recognition chip that uses a fraction of the power of existing technologies"},"content":{"rendered":"<p><img decoding=\"async\" src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=738\" class=\"\"\/><\/p>\n<p>MIT announced today that it\u2019s developed <a target=\"_blank\" href=\"http:\/\/news.mit.edu\/2017\/low-power-chip-speech-recognition-electronics-0213\">a speech recognition chip<\/a> capable of real world power savings of between 90 and 99 percent over existing technologies. Voice technology\u00a0has, of course, become nearly ubiquitous in mobile devices, thanks to the exponential growth of smart assistants like Siri, Alexa and Google Home \u2013 but the new chip could help branch out in much simpler electronics.<\/p>\n<p>The team gives\u00a0IoT devices a potential use case\u00a0\u2013 devices designed to go months on end without charging or changing batteries. Here\u2019s MIT professor Anantha Chandrakasan on the chip,<\/p>\n<blockquote readability=\"8\">\n<p>Speech input will become a natural interface for many wearable applications and intelligent devices. The miniaturization of these devices will require a different interface than touch or keyboard. It will be critical to embed the speech functionality locally to save system energy consumption compared to performing this operation in the cloud.<\/p>\n<\/blockquote>\n<p>The technology features a \u201cvoice activity detection\u201d circuit capable of distinguishing ambient noise from speech, turning on on-board speech recognition hardware when it detects the latter.<\/p>\n<p class=\"p1\"><span class=\"s1\">Michael Price, a graduate student who worked on the project, gave TechCrunch a bit more detail regarding the system\u2019s built-in speech detection,<\/span><\/p>\n<blockquote readability=\"14\">\n<p class=\"p1\"><span class=\"s1\">The chip that we demonstrated includes a continuous speech recognizer based on hidden Markov Models (HMMs).\u00a0 It transcribes an arbitrary length audio input into a sentence.\u00a0 The transition model is a weighted finite-state transducer (WFST).\u00a0 The acoustic model is a feed-forward neural network.\u00a0 The same general techniques are used in some software speech recognizers.<\/span><\/p>\n<p class=\"p1\"><span class=\"s1\">We trained models for this recognizer using Kaldi, an open source toolkit.\u00a0 We used a few different speech datasets for training and testing.\u00a0 The largest recognizer we tested had a vocabulary of 145k words and required 7.78 mW for real-time operation.\u00a0 The smallest was a digit recognizer (11 words including \u201coh\u201d for zero) which required 172 uW.<\/span><\/p>\n<\/blockquote>\n<p>The chip is essentially designed to be always on in a low-power mode, switching over when voice is detected, thus making it ideal for technologies like wearable devices, which can benefit from speech control, but are required to last much longer\u00a0on a single charge than a standard handset.<\/p>\n","protected":false},"excerpt":{"rendered":"<p><img loading=\"lazy\" decoding=\"async\" width=\"680\" height=\"383\" src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=680\" class=\"attachment-large size-large wp-post-image\" alt=\"google-voice-android\" srcset=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=680 680w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=1360 1360w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=150 150w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=300 300w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/01\/google-voice-android.jpg?w=768 768w\" sizes=\"(max-width: 680px) 100vw, 680px\" \/>&nbsp;MIT announced today that it&#8217;s developed a speech recognition chip capable of real world power savings of between 90 and 99 percent over existing technologies. Voice technology\u00a0has, of course, become nearly ubiquitous in mobile devices, thanks to the exponential growth of smart assistants like Siri, Alexa and Google Home &#8211; but the new chip could help branch out in much simpler&#8230; <a href=\"https:\/\/techcrunch.com\/2017\/02\/13\/mit-speech-chip\/?ncid=rss\">Read More<\/a><\/p>\n<div class=\"feedflare\">\n<a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:2mJPEYqXBVI\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=2mJPEYqXBVI\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:7Q72WNTAKBA\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=7Q72WNTAKBA\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:yIl2AUoC8zA\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=yIl2AUoC8zA\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:-BTjWOF_DHI\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?i=6MkSNPeUvMQ:4Eq_XO68cOA:-BTjWOF_DHI\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:D7DqB2pKExk\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?i=6MkSNPeUvMQ:4Eq_XO68cOA:D7DqB2pKExk\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=6MkSNPeUvMQ:4Eq_XO68cOA:qj6IDK7rITs\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=qj6IDK7rITs\" border=\"0\"><\/img><\/a>\n<\/div>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~r\/Techcrunch\/~4\/6MkSNPeUvMQ\" height=\"1\" width=\"1\" alt=\"\"\/> <\/p>\n","protected":false},"author":5817,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"categories":[25],"tags":[59,66,26],"class_list":["post-333239","post","type-post","status-publish","format-standard","hentry","category-technology","tag-media","tag-techcrunch","tag-technology"],"_links":{"self":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/333239","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/users\/5817"}],"replies":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/comments?post=333239"}],"version-history":[{"count":0,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/333239\/revisions"}],"wp:attachment":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/media?parent=333239"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/categories?post=333239"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/tags?post=333239"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}