I have recently started searching for an application to help me do some daily training. After trying some of them I have realised that you have to interact with the application after each exercise and that the sets of exercises are predefined. Therefore, I have thought of implementing an application which you can speak to, such that it will suggest you your desired type of exercise. Besides that, the music during your training session plays an important role as well. So, we have implemented an innovative way of choosing the right music for you.
What it does
First of all, after the user says how he feels, our application does a speech recognition and then NLP Sentiment Analysis, in order to start the music that fits your mood best. The next step is to mention what would you like to train more exactly. After a speech recognition and a NLP algorithm, the desired exercise starts together with the music we have already selected at the previous step.
How we built it
The back-end is written totally in Python. We are using the Flask framework, NLPToolkit library and Uberi for speech recognition. The front-end was built in foundation, sass, css, angularjs, jquery.
Challenges we ran into
Time management was a problem and we should also mention the browser and platforms' compatibilites.
Accomplishments that we're proud of
A good looking and functional project.
What we learned
flask flask flask flask
What's next for Personal AI Trainer
We have thought of a Machine Learning algorithm , that predicts the days in which the user is most likely to skip his training. Therefore, we will be using Twilio to send a motivational text message. We have implemented the Twilio's API side, however we still have to work on the Machine Learning side.