# Title Team Members TA Documents Sponsor
54 Music Discovery Band
Michael Faitz
Nitin Jaison
Vignesh Srivatsan
Stephanie Jaster design_document1.pdf
Michael Faitz (mfaitz2), Nitin Jaison (jaison2), Vignesh Srivatsan (vsrvtsn2)
Music Discovery Band

Problem - We want to simplify the seemingly random task of discovering new music for Spotify users. Many people do not know how to go about finding new music because they don’t know if they will like it in the moment. There currently exists no technology that can provide song suggestions from the entire Spotify library by keeping track of a person’s physical activity levels, environment, and even mood. We hope to use this information to match users to songs that they would be more likely to listen to while they undergo any situation of varying intensity. For example, we would match fast-paced, vibrant songs for periods when the user does physical activity, or calmer, softer songs to accompany the user doing leisurely activities around the home.

Solution Overview - We propose a wearable wristband that can sense a user’s activity level and environment so that this information can be sent to a mobile application to determine ideal music to listen to. When determining what type of music might best accompany any given situation, we must first determine the different motivating factors for listening to music. We mentioned earlier that we aim to match songs of high intensity to the user when they undergo physical stress on the body, like during exercise. For this reason, the wristband will include both a heart beat sensor and an accelerometer in order to keep track of the user’s physical stats over an extended period of time and suggest music of similar intensity levels. Another motivating factor to determining music is the user’s environment, as the type of music they want to listen to might change in different surroundings. For this reason, we also intend to include an ambient sound sensor that can detect noise level changes in the user’s environments to provide further information to the song selection process. Finally, a person’s mood has a significant impact on the music they listen to, and we aim to account for that by implementing physical buttons on the device that the user can press to indicate different moods, which will in turn add a “filter” to the type of music that will be played to more closely match the user’s mood. A bluetooth transmitter will be used to connect the wristband to the user’s phone, where information regarding the user’s heart rate, acceleration, level of noise, and mood will be used to queue up songs to listen to. The app will use this information to find songs that match the user’s situation by finding songs through attributes such as tempo, loudness, and energy. By using the entire Spotify library we give the ability to discover new music, filtering through to find certain songs that correspond with the information from the wristband. We also understand that different people’s heart rates means different things, so we will set a baseline resting, light activity, and rigorous activity heart rate when the user first uses the app. We can do this by asking the user to rest, do a light jog, and run, collecting their heart rate for each activity.

Solution Components:
[Subsystem #1] : Sensors
- Since the goal of the device is to capture as many aspects of the individual’s life as possible to make an informed decision about music, there are several sensors that we wish to implement into the wristband. The general layout of the wristband will have the PCB and a heart rate sensor on the bottom side of the wrist to combine the flat zone needed for the PCB within the wristband with the location that the heart rate sensor needs to be for the user. This will provide clear indication to the user that the wristband is on correctly as well as effectively simplifying our layout. We also wish to have an ambient noise monitor within the design to tell us information about the current situation of the user whether they are at home or in a more noisy public environment. Another sensor we want to add is an accelerometer which can read the users current speed and use that to consider the type of music they may want to listen to. If a person is running and moving fast they may want to listen to more upbeat music whereas sitting down at the computer or to read would dictate a more peaceful/calm song choice. Finally, we want to implement some selection of moods available for the user so that they can further provide a filter for the song discovering depending on how they are feeling at the time. These inputs will take the form of small buttons available on the wristband which indicate moods such as “Relaxed”, “Working Hard”, “Frustrated” and factor them into our selection. Along with these buttons will be options for stopping the song, skipping the song, and volume control.
- [Subsystem #2] Power:
- The wristband will need some compact power supply to make all other subsystems function. For this a lithium battery should be sufficient due to its relatively low size and considerable power output.
- [Subsystem #3] - Software:
- There will be a mobile application that will be connected to a user’s Spotify account and the wristband through the phone’s bluetooth capability. After setting the baseline heart rates when the app is first used, it will receive heart rate, acceleration, outside sound, and mood information from the wristband. An algorithm will then be used to filter through Spotify’s music library and match songs using Spotify’s audio analysis features to find ones that match the activity level, environment, and mood of the user. Before the current song ends the app will use the updated information from the wristband to queue up a new song.
- [Subsystem #4] - Control:
- The control for this wristband can be handled via a microprocessor with the PCB and will process information collected from the various sensors and user inputs and send the processed data discussed in the software subsystem to a bluetooth transmitting device that will be connected to the users phone.

Criterion for Success - Describe high level goals of what your project needs to meet to be effective.
- Wristband with accelerometer, heart-beat sensor, sound sensor, and mood buttons that can accurately track information as its pertains to the user’s activity level, surrounding environment, and mood.
- The wristband should also be able to transmit the information it collects from its sensors and buttons to the smartphone app through a Bluetooth transmitter
- The smartphone app must then take this information and use it to automatically queue up and play an “appropriate” song from Spotify’s library. A song is deemed appropriate by an algorithm that we will implement that will consider all of the factors listed above given from the information transmitted by the wristband. It will use this information to queue up a song that should match the user’s activity level and mood.

Low Cost Myoelectric Prosthetic Hand

Michael Fatina, Jonathan Pan-Doh, Edward Wu

Low Cost Myoelectric Prosthetic Hand

Featured Project

According to the WHO, 80% of amputees are in developing nations, and less than 3% of that 80% have access to rehabilitative care. In a study by Heidi Witteveen, “the lack of sensory feedback was indicated as one of the major factors of prosthesis abandonment.” A low cost myoelectric prosthetic hand interfaced with a sensory substitution system returns functionality, increases the availability to amputees, and provides users with sensory feedback.

We will work with Aadeel Akhtar to develop a new iteration of his open source, low cost, myoelectric prosthetic hand. The current revision uses eight EMG channels, with sensors placed on the residual limb. A microcontroller communicates with an ADC, runs a classifier to determine the user’s type of grip, and controls motors in the hand achieving desired grips at predetermined velocities.

As requested by Aadeel, the socket and hand will operate independently using separate microcontrollers and interface with each other, providing modularity and customizability. The microcontroller in the socket will interface with the ADC and run the grip classifier, which will be expanded so finger velocities correspond to the amplitude of the user’s muscle activity. The hand microcontroller controls the motors and receives grip and velocity commands. Contact reflexes will be added via pressure sensors in fingertips, adjusting grip strength and velocity. The hand microcontroller will interface with existing sensory substitution systems using the pressure sensors. A PCB with a custom motor controller will fit inside the palm of the hand, and interface with the hand microcontroller.

Project Videos