東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Low Power Audio Feature Extraction for Machine Learning Applications.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Low Power Audio Feature Extraction for Machine Learning Applications./
作者:	Villamizar, Daniel Augusto.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
面頁冊數:	103 p.
附註:	Source: Dissertations Abstracts International, Volume: 83-05, Section: B.
Contained By:	Dissertations Abstracts International83-05B.
標題:	Deep learning. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28812985
ISBN:	9798494463272

Low Power Audio Feature Extraction for Machine Learning Applications.
Villamizar, Daniel Augusto.

Low Power Audio Feature Extraction for Machine Learning Applications. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 103 p.

Source: Dissertations Abstracts International, Volume: 83-05, Section: B.

Thesis (Ph.D.)--Stanford University, 2021.

This item must not be sold to any third party vendors.

Always-on sound classification is a desirable but power-intensive function for a variety of emerging applications such as wearables and IoT devices. The hardware energy consumption of sound classifiers is typically driven by signal digitization, feature extraction, classification model storage, and execution. Yet, the complexity of the signal of interest is often much lower than that of the raw signal acquired from the microphone and processed by a machine learning engine. For instance, semi-stationary sounds (e.g., engine noise, baby cry, running water, human chatter, etc.) are signals with lower information content than more complex sounds such as music or speech. In this dissertation, I will present the benefits of leveraging an engineered feature set to efficiently classify semi-stationary sounds. This approach requires one to three orders of magnitude fewer parameters and can be therefore trained over ten times faster than competitive deep learning models. I will also describe a circuit topology and system architecture that can be used to extract both engineered features as well as more general purpose ones. Our work resulted in a 32-channel analog filterbank IC for audio front-end signal processing. It employs a passive N-path switched capacitor topology to achieve high power efficiency and reconfigurability. The circuit's unwanted harmonic mixing products are absorbed by the machine learning model during training. To enable a systematic pre-silicon study of this effect, we develop a computationally efficient circuit model that can process large machine learning datasets in practical run-times. Measured results using a 130 nm CMOS prototype IC indicate competitive classification accuracy on datasets for baby cry detection (93.7% AUC) and voice commands (92.4% average precision), while lowering the feature extraction energy compared to digital realizations by approximately 2x and 10x, respectively. The 1.44 mm2 chip consumes 800 nW, which corresponds to the lowest normalized power per simultaneously sampled channel in recent literature.

ISBN: 9798494463272Subjects--Topical Terms:

3554982
Deep learning.

Low Power Audio Feature Extraction for Machine Learning Applications.
LDR:03269nmm a2200397 4500 001 2344696
005 20220531064628.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798494463272
035 $a (MiAaPQ)AAI28812985
035 $a (MiAaPQ)STANFORDxf872vs2626
035 $a AAI28812985
040 $a MiAaPQ $c MiAaPQ
100 1 $a Villamizar, Daniel Augusto. $3 3683494
245 1 0 $a Low Power Audio Feature Extraction for Machine Learning Applications.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 103 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-05, Section: B.
500 $a Advisor: Murmann, Boris;Raina, Priyanka;Rivas-Davila, Juan.
502 $a Thesis (Ph.D.)--Stanford University, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a Always-on sound classification is a desirable but power-intensive function for a variety of emerging applications such as wearables and IoT devices. The hardware energy consumption of sound classifiers is typically driven by signal digitization, feature extraction, classification model storage, and execution. Yet, the complexity of the signal of interest is often much lower than that of the raw signal acquired from the microphone and processed by a machine learning engine. For instance, semi-stationary sounds (e.g., engine noise, baby cry, running water, human chatter, etc.) are signals with lower information content than more complex sounds such as music or speech. In this dissertation, I will present the benefits of leveraging an engineered feature set to efficiently classify semi-stationary sounds. This approach requires one to three orders of magnitude fewer parameters and can be therefore trained over ten times faster than competitive deep learning models. I will also describe a circuit topology and system architecture that can be used to extract both engineered features as well as more general purpose ones. Our work resulted in a 32-channel analog filterbank IC for audio front-end signal processing. It employs a passive N-path switched capacitor topology to achieve high power efficiency and reconfigurability. The circuit's unwanted harmonic mixing products are absorbed by the machine learning model during training. To enable a systematic pre-silicon study of this effect, we develop a computationally efficient circuit model that can process large machine learning datasets in practical run-times. Measured results using a 130 nm CMOS prototype IC indicate competitive classification accuracy on datasets for baby cry detection (93.7% AUC) and voice commands (92.4% average precision), while lowering the feature extraction energy compared to digital realizations by approximately 2x and 10x, respectively. The 1.44 mm2 chip consumes 800 nW, which corresponds to the lowest normalized power per simultaneously sampled channel in recent literature.
590 $a School code: 0212.
650 4 $a Deep learning. $3 3554982
650 4 $a Human performance. $3 3562051
650 4 $a Christianity. $3 581949
650 4 $a Signal processing. $3 533904
650 4 $a Internet of Things. $3 3538511
650 4 $a Families & family life. $3 3422406
650 4 $a Design. $3 518875
650 4 $a Breakdowns. $3 3682712
650 4 $a Energy. $3 876794
650 4 $a Algorithms. $3 536374
650 4 $a Keywords. $3 3560140
650 4 $a Sound. $3 542298
650 4 $a Acoustics. $3 879105
650 4 $a Artificial intelligence. $3 516317
650 4 $a Computer science. $3 523869
650 4 $a Electrical engineering. $3 649834
650 4 $a Engineering. $3 586835
650 4 $a Individual & family studies. $3 2122770
650 4 $a Religion. $3 516493
690 $a 0791
690 $a 0389
690 $a 0986
690 $a 0800
690 $a 0984
690 $a 0544
690 $a 0537
690 $a 0628
690 $a 0318
710 2 $a Stanford University. $3 754827
773 0 $t Dissertations Abstracts International $g 83-05B.
790 $a 0212
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28812985