TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Signal processing is key to embedded machine learning

104 pointsby janjongboomabout 5 years ago

6 comments

alephnilabout 5 years ago
Actually signal processing is already used for most machine learning of audio signals, including speech recognition. The reason is that ML algorithms, including deep learning has a hard time learning the information you can get from a discrete Fourier transform.<p>Audio data in time domain are just too noisy for most machine learning, and doing some signal processing as a preprocessor step often helps a lot.<p>Here it seems like he works with non-audio data, where this is less common.
评论 #22648788 未加载
kbumsikabout 5 years ago
It&#x27;s not the key but the fundamentals that we have done for decades...
proverbialbunnyabout 5 years ago
This is an awesome topic, but I&#x27;m somewhat annoyed they didn&#x27;t dive into what kind of DSP and instead turned the article into an advertisement.<p>Does anyone have any good further reading on the topic? (Books, articles, classes, anything really.)
评论 #22649027 未加载
评论 #22657055 未加载
评论 #22651256 未加载
ganzuulabout 5 years ago
The Kalman filter is basically an ML algo. The key here is to implement already known linear optimization approximation versions of it in common libraries.
评论 #22649508 未加载
uoaeiabout 5 years ago
It always comes down to representation. If you can use a deterministic, efficient algorithm to represent the data in a more amenable manner, then the ML system will have a much easier time &quot;making sense&quot; of the patterns inherent in the data compared to a system that has to learn some abstract transformation from raw data to useful representations.
taericabout 5 years ago
My concern with a lot of signal processing techniques used in ml is that sometimes they presuppose things that may not be true.<p>That is, signal processing had Nyquist&#x27;s rates. And typically knows there is an underlying signal. Does ml have either?
评论 #22649167 未加载
评论 #22647634 未加载
评论 #22647516 未加载