Scribd
Upload
13 views

TinyML App

The document discusses developing a tiny machine learning framework and application for an embedded device without connectivity. The tasks involve defining an application use case, studying an STM32F407 device and CMSIS APIs, developing framework APIs, and selecting applications like predicting failures from sensor data or removing noise from phone calls.

Uploaded by

Arjun achar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
13 views

TinyML App

The document discusses developing a tiny machine learning framework and application for an embedded device without connectivity. The tasks involve defining an application use case, studying an STM32F407 device and CMSIS APIs, developing framework APIs, and selecting applications like predicting failures from sensor data or removing noise from phone calls.

Uploaded by

Arjun achar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 4

TinyML_EmbeddedApp

Project Briefing

Venkatarangan MJ
Department of Electrical and Electronics Engineering
Problem Statement/Work Scope

• Explore ML running on Embedded Device with constrains and without


any connectivity
• Prove the possibilities of TinyML with framework development for
reference applications
• If needed, add any needed algorithms for embedded world

• Develop a Tiny ML framework and App for embedded device


• Platform to be explored: CMSIS (generic API for ARM controllers)
• App to be selected such that data can be generated easily
• Application possibilities (in discussion with known companies)
Tasks involved

• Define the Application and use cases (trying to look for open problem statements
and specific ones from companies)
• Study STM32F407 embedded device and CMSIS APIs for machine learning
• Develop the needed framework APIs for applications and specific application to
deploy
• Applications to be selected:
• Analyze sensor data from any industrial equipment to predict the failure ->
Prevents down time
• Smart Devices identify customer usage and recommend customised and context
based inputs
• Mobile phone removing background noise when some-one talks
(sender/receiver side)
• Smart wearables: Monitor health data and provide personalized real-time
feedback without connectivity.
• Analyse visual data from cameras to automate process and quality control
THANK YOU

Venkatarangan MJ
Department of Electrical and Electronics Engineering
[email protected]
+91 80 6666 3333 Extn 876

You might also like