What are the most significant risks that could jeopardize the success of the project? How are these risks being managed? What contingency plans are ready?
Our biggest risks at the moment are coming from our signal processing for scoring and from our ability to do vocal removal with low latency. Scoring is risky because as we have recently moved away from pitch detection, we are now developing new ideas for metrics and if we are not able to select one and start testing it as soon as possible, we leave a risk of missing out on a major component of our gamification. For our vocal removal, we have confirmed that it is possible with low latency via software, but we still need this to become a hardware system ideally. Both risks are being managed by quick decision making, as we are finalizing design choices right now and are hoping to be able to quickly get to a testing phase and prototype these pieces as soon as possible. As far as contingency, for our scoring, we have a range of options including some extremely easy (but unideal) work arounds. For the vocal removal, there is always the option to use an AI model to do the work which is proven to be possible.
Were any changes made to the existing design of the system (requirements, block diagram, system spec, etc)? Why was this change necessary, what costs does the change incur, and how will these costs be mitigated going forward?
The primary change is switching from a bandpass / bandstop filter to the subtraction method taking advantage of stereo output. There is not much as far as surface level cost, but this does limit our potential music library to only songs that exist in stereo format. It will simply require a small check with the Spotify API to verify before allowing the user to confirm.
Additional Questions: