Development of a simple Hidden Markov Model (HMM) for speech tagging.
The notebook will guide the reader through the code needed for speech tagging, developing a model achieving over 95% accuracy on speech tagging tasks.
- Clone the repository on your computer. For those who are not familiar with command line utilities, GitHub has developed this utility which lets do pretty much everything you need through a GUI. GitHub Desktop
- Open a terminal window and navigate to the project folder. Open the notebook HMM_tagger.ipynb and follow the instructions.
NOTE: Chances are part of the code may not run due to some missing packages. Please make sure to go through the notebook and retrieve all the necessary packages. The list can be found in requirements.txt.
Sometimes GitHub cannot render the jupyter notebook properly. In such a situation, go to the nbviewer website and copy and paste the URL address of the notebook so as to render it on the browser.