https://github.com/tum-ei-eda/micro-kws
Complete flow for keyword spotting on microcontrollers. From data collection to data preparation to training and deployment.
https://github.com/tum-ei-eda/micro-kws
Last synced: 2 months ago
JSON representation
Complete flow for keyword spotting on microcontrollers. From data collection to data preparation to training and deployment.
- Host: GitHub
- URL: https://github.com/tum-ei-eda/micro-kws
- Owner: tum-ei-eda
- Created: 2022-07-14T06:40:25.000Z (over 3 years ago)
- Default Branch: workshop
- Last Pushed: 2025-03-10T12:24:37.000Z (8 months ago)
- Last Synced: 2025-09-01T16:58:15.500Z (2 months ago)
- Language: Jupyter Notebook
- Size: 21.8 MB
- Stars: 4
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Micro KWS
Complete flow for keyword spotting on microcontrollers. From data collection to data preparation to training and deployment.
## Context
This project is used as part of the lab accompanying the lecture: Embedded System Design for Machine Learning offered by EDA@TUM.
## Structure of this repository
The following directories can be found at the top level of this repository:
- `record/`: Provides utilities for recording and preprocessing new dataset samples (Optional)
- `train/`: Contains MicroKWS training flow and tutorial (Lab 1)
- `tvm/`: Contains a tutorial for generating MicroKWS kernels for a pre-trained model using the TVM Framework (Lab 2, Part 1)
- `target/`: Provides target software demo for deploying the MicroKWS application to a microcontroller (Lab 2, Part 2)
- `debug/`: Contains a python tool to debug the target application running on the device (Lab 2, optional)