After an intense couple of weeks of preparation, Vitalis participated in IPPD’s Prototype Inspection Day event in the Rion Ballroom last Tuesday. Upon activating our prototype, a recording session from our edge device was able to begin collecting audio data. After recording, our prototype was able to send it to an external server (Matthew’s Laptop) for transcription synthesis upon conclusion. One finished the transcript was sent back and was able to be read from the prototype’s monitor.

Here is a picture of our edge device.

The whole team was happy to receive feedback on our prototype from faculty working in AI and BME. The input we were given gave us new considerations to be conscious about moving into our next major event, SLDR.

Here is a picture of the whole group at Prototype Inspection Day. From left to right we have Aizlyn Potter (ME), Huy Nguyen (CpE), Julia Leser (BME), Julianna Gonzalez (BME), Daniel Saurez (CS), Ezequiel Medrano (CS), Dr. Laura Cruz Castro (Vitalis coach), and Matthew Li (CS).

Moving forward, in addition to preparing for SLDR, We are all very excited to begin adding more components that will be present in our final design. we hope to deliver a new chassis, begin working on our GUI, decide on and implement an E2E encryption method, and bring our team to UF’s Center for Experiential Learning and Simulation Lab to gain a better understanding of ICU environments and code blue calls.

Team Vitalis had a busy week!

We first presented for UF AI Days – which occurred on Tuesday Oct 28th from 5:00-6:00 PM. In order to adequately prepare, we created a Presentation poster which required input from our liaisons as to what could be publically presented from our NDA. We also prepared speaking parts for each member for the actual presentation!

Here is a photo that we have from the event!

We’ve also been able to make progress on prototyping from a software and hardware side. We’ve been able to begin utilizing the processor in conjunction with the purchased microphones as well as testing external server API calls with AI model transcriptions!
Currently we are preparing for our Prototype Inspection Day on November 4th, and will provide update after the event with any feedback!

Very exciting developments were made this week from everyone on team Vitalis! Our software team has developed a preliminary partioning system to break up recorded data into smaller pieces for more complete recording and processing. Additionally, we have selected a preliminary model for summarization of the recorded material, which we will begin testing soon!

Additionally, our Hardware sub-team has completed a preliminary design for the prototype chassis to be presented on Prototype Inspection Day (11/04). While very barebones at the moment, we plan to make vast improvements based on the feedback we recieve from professionals, students and stakeholders who will be examining our design in depth, among those of all the other IPPD teams!

The team visited JWM Engineering this Friday and met with the JWME team! During this visit, Vitalis had the opportunity to present the Code Blue Preliminary Design Review to Stephen and Roseann Belser. The team also toured the office and ate a wonderful lunch together. The team was able to see a JWME server that could potentially provide support to our project!

Now that PDR is over, Vitalis is extremely excited to start prototyping! The software team is starting to design more audio experiments and brainstorming details on how to use a server to process our various models. Additionally, the hardware team is planning to start testing the capabilities of a Raspberry pi 4 and 5 with our current models. Let’s get to work!

This week was Team Vitalis’s first go at a full team presentation. On Tuesday afternoon (October 7th), Team Vitalis presented their shortened PDR presentation for the first time to three other teams, multiple different coaches, and IPPD staff.

Formally, Team Vitalis provided the PDR Peer Review committee a status update on the Code Blue project on 7 October 2025. Attending faculty, staff, and invited guests included Dr. Laura Cruz Castro, Dr. Janise McNair, Dr. Christian Grant, and Arielle Barnard.

Overall, we received amazing feedback from all parties involved, and it mainly focused on two points:

1. Presentation Skills
   • Our team did a wonderful job with most presentation scores being Competent or Excellent! We need to make sure to support one another as we present which includes looking at our speaker, nodding along with them, and smiling and being understanding towards the crowd. We also need to make sure we project our voices and bring plenty of energy to keep the attention of the listener and also showcase our passion for the project!
2. Slide Composition
   • For some of our slides (specifically the House of Quality and Project Risk Table), the listeners let us know that it was harder to understand the image because of the amount of words on the screen as well as the small font. Because of this, we’ve separated those slides to make it easier on the listener to understand and read during the presentation. Here is one of the slides pictured below:

This week was a very busy week for team Vitalis! The majority of the focus was on refining the Product Architecture and developing models for our ideas created in the concept generation phase. Ezequiel developed some very elaborate and informative flow charts of the final product that we plan to continue updating as more research is done. Alongside this, Huy developed some 3D models based on Ezequiel’s plans. Below are photos of each!

This week, Vitalis had no shortage of assignments and research to do on the subject. Rigorous research, deadlines regarding our product design documentation, preparing for the career fair, and a challenging start to midterm season for our team. Nevertheless, Vitalis powered through and has made some big steps this week!

During the prior weekend, Daniel got the ball rolling for what will be the foundation for our Automatic Speech Recognition (ASR) engine. Utilizing Open AI’s Whisper, Daniel created a small program that takes in audio data, converts it to a wav file, and runs it through a pre-selected Whisper model.

Eagerly, Vitalis began conducting small tests on the effectiveness on three different Whisper models by measuring word error rate (WER). We benchmarked the models in settings of Marston ambiance and isolated silence to gauge the efficacy of the specific model and microphone in real world environments.

Additionally, we began creating flow diagrams to itemize our software components to figure out how many AI models we will need in addition to statically-coded components. This aims to make it easier to allocate responsibilities during development and to make our design modular so components can be easily removed and added.

Lastly, our liaisons and coach, during this weeks meeting, were very impressed with our progress and we are perpetually excited to keep this momentum going forward into the stages of designing and prototyping.

Week 4 was a busy week for Vitalis! Some of our team members attended a Team Building Workshop hosted by Florida Power and Light, ending up with White Belts in the Six Sigma methodology. The team worked hard on developing a project roadmap and considering the many risk factors for our product. We also did a lot of brainstorming as to what our project could look like! Here’s a picture of us with one of our favorite ideas!