We (Franziska and Mo) wanted to have a fitness coach which 1) Ran exercises at our pace (slow, fast or variable speed) 2) Was not voice only as we like to see images or videos and be able to link more data to it from our devices 3) Talk us through the routine so we know what we are doing 4) Could be managed by us not us being managed by it 5) Had a large library of routines 6) Had music associated with each routine
But discovered that it didnt exist. So we decided to start building one for own use.
What it does
It starts off by learning a little about you and what you like to do, but leaves the control in your hands. You are always in charge. It has several set routines or you can jump to any one that you prefer. It shows you the routine on the screen (TV, Echo Show) as well as talking you through it It starts the routine but allows you to slow it down or speed it up or leave and come back to it
How I built it
We did a basic user research and wrote a few user stories. We then created an empathy map followed by user flows. We designed them in Adobe XD and Figma. We created a series of graphical assets and audio clips to go with the skill and then designed the experience in Voiceflow.
The process was very detailed as we are looking to continue building on this activity to improve the app.
Challenges I ran into
Designing for voice and screen is not very straightforward. The user flows started to become complex requiring detailed testing (which took a long time). Synchronising images, audio and speech to create a meaningful experience requires attention to details. As a team we had to continiously be in touch.
Accomplishments that I'm proud of
Pulling all the resources together in order to deliver the multimodal experience was a great and satisfying experience. We managed to create an app which uses a combination audio, speech and display devices to deliver an experience that looks and feels good. We believe that APL and multi-modality will change the way people use Alexa and will proliferate across many areas specially in consumer and business.
What I learned
Improve the way we create voice apps Learn to use images and music to enhance the voice experience Working better as a team on voice based application
What's next for Fit Away
Following the initial testing, we like to do further tests on FireTV, expand the library of routines and the graphics library, bring in animation and video routines, , extend the number of intents we deal with, add additional music to the library