Piano Player & Learner

A digital piano designed to help users learn and play songs with on-screen note guidance.

info

Author: Sirbu-Boeti Eduard-Cristian
GitHub Project Link: https://github.com/UPB-PMRust-Students/project-dm-2025-Sirbu-Boeti-Eduard

Description

The project is a 7-note piano intended as a simple learning system to follow the notes of a song. The user can play individual notes or up to two notes simultaneously, allowing for basic chords and harmonies aimed towards beginners. For intermediate learners we will make use of a joystick to experiment with changing the pitch of a note.

• We will use an LCD display to:
  • Guide the user to follow the displayed notes of a chosen song
  • Indicate system states (e.g. playing, recording, name of the song to be played etc.)
• The interface is a 4x4 button matrix used for controlling the system:
  • 1st row: Play more difficult pre-recorded songs
  • 2nd row: Guided play-along mode for easier songs using the display
  • 3rd row: Record and save user-played songs
  • 4th row: Auxiliary actions (start recording, confirm, cancel, etc.)

Motivation

The main motiviation behind this project was me wanting to make use of the 4x4 button matrix component. Initially I wanted to make the tabletop game Battleship, but I wasn't satisfied with there being only 16 possible square grids due to the size of the button matrix.

After fiddling around with buttons on the breadboard I liked the feel of it and wanted to try doing something with buttons that will fit nicely in your hand. Due to the size of the breadboard I was thinking of maybe using it as a remote controller for something but I ended up going with a small 7-note piano.

This was due to the fact that, when I was younger, I used to have a Yamaha digital piano with a lot of similar features and also because I saw that in previous years there have been similar piano projects.

Architecture

Inputs

• 7× Push Buttons
  • Act as the main “piano keys” (Do, Re, Mi, Fa, Sol, La, Si)
  • Connected as GPIO inputs with pull-ups
• 4×4 Button Matrix (Keypad)
  • Used for system control (mode selection, song selection, confirm/cancel actions)
  • Connected via GPIO row/column scanning
• Analog Joystick
  • VRx and VRy connected to ADC inputs
  • Used for experimenting with pitch control
  • The joystick push button (SW) is not used/needed

Outputs

• 2× Passive Buzzers
  • Controlled using PWM
  • Support up to two simultaneous notes, naming the buzzers as "voices"
• 16×2 Character LCD (HD44780 controller chip)
  • Displays current mode, target notes and system status
  • Connected via I2C

Connections

• GPIO / EXTI – push buttons (note input)
• GPIO scan – 4×4 keypad (columns driven, rows read)
• ADC – joystick VRx / VRy
• PWM – passive buzzers (frequency = note pitch)
• I2C – LCD module

Log

Week 5: 27 October - 2 November

This was the week when we had to present our Powerpoint presentations. In this week, I just gathered the necessary information and created the presentation. There were no additional modifications requested by the professor, as such I took that the project received the "OK".

Week 6: 3 November - 9 November

This week I set up all 7 buttons on the breadboard and starting building a quick demo to see the overall difficulty. Initially I used a single buzzer, but I also wanted an additional one for concurrent notes. I didn't have an additional passive buzzer (only an active one) so I had to order one.

While waiting for the new buzzer, I managed to create the functional demo where i could play the notes through the buzzer.

Week 7: 10 November - 16 November

This week I wanted to connect the I2C display to the MCU. I initially had trouble with the identification of the I2C address, thus not being able to use the display and, as such, I had to wait to complete Lab 06 - "Inter-Intergrated Circuit" to finally be able to scan the correct I2C address and print "Hello World".

Around this time, I also received my second passive buzzer and I finally connected it and was able to play two notes at the same time.

Side note: I had to change all the button and the existing buzzer connections because I had to get the two buzzers to connect on different TIMs in order to play different notes at the same time.

Week 8: 17 November - 23 November

At this point, I had all the buttons and buzzers ready and the LCD display was just connected. Now, I wanted to connect the 4x4 button matrix. Here I struggled a lot because there were no markings on the on PCB of the order of the pins and their correspondence with the rows and columns. After finally thinking I got the connections right, I observed that it still wasn't working. I found out the the button matrix that I had simply wasn't detecting the entire rows 1 and 2 (I assume the pins that denoted the rows were broken because the columns had not problems). As such, I postponed this step in order to buy a new button matrix that hopefully works.

Side note: At this point, I deconnected for now the existing note buttons for testing as I no longer had space for the 8 pins of the button matrix and had no Female-to-Female connectors for connecting to the Morpho pins.

Week 9: 24 November - 30 November

This week I started testing my joystick to see if it is working properly. I decided that I didn't need to make use of the SW button (clicking the joystick) and that I will either use the X or Y axis for pitch control. After testing it worked properly the ADC and seeing the changes in terminal. At this point I wanted to try testing the pitch change on a static note, but the results were not satisfying, making it sound bad. Hoping to find a solution.

Week 10: 1 December - 7 December

This week I mostly did changes to the code in order to better segment the code for future implementation. Namely, preparing the "Free Play", "Play Along 1 (single notes)" and "Play Along 2 (double notes)" into different functions to be called in the "loop".

Week 11: 8 December - 14 December

After buying some Female-to-Female cable I was able to complete connection of all components together. I ended up not ordering a new button matrix, thinking that maybe it was my code's fault it wasn't working, but after several more tries I concluded that I need to buy a new button matrix. For now, until I buy the new one, I will have to work on the currently functional rows.

Week 12: 15 December - 21 December

This week I worked on writing the documentation and moving my segmented demos into a single Github repository for the source code. Starting to focus on the source code with all of its features.

Hardware

I have provided links for reference.

• STM32 NUCLEO-U545RE-Q

• Passive buzzers (×2)

• Push buttons (×7)

• 16×2 character LCD display

• 4×4 button matrix

• Analog 2-axis XY joystick

• Connecting components:

  • Breadboard
  • Male-to-Male, Male-to-Female, Female-to-Female Cables

Current pin usage

• LCD (I2C1)
  • SCL: PB6
  • SDA: PB7
• Buzzers (PWM)
  • Buzzer 1: PC9 on TIM3_CH4
  • Buzzer 2: PB10 on TIM2_CH3
• 7 Push Buttons
  • PC8, PB3, PB5, PB4, PA2, PC7, PC6
• 4×4 Button Matrix (Keypad)
  • Columns 1->4: PC13, PC12, PC11, PC10
  • Rows 1->4: PA9, PA10, PA11, PA12
• Joystick
  • VRx / VRy on A0, A1
  • powered at 3.3 V
  • Not using the SW(Joystick Click)

Schematics

Bill of Materials

Device	Usage	Quantity	Unit Price	Subtotal
STM32 NUCLEO-U545RE-Q	Main microcontroller board; handles all input processing, sound generation and display control	1	~100 RON	~100.00 RON
Passive buzzer 3V	Audio output for generating musical notes via PWM	2	0.99 RON	1.98 RON
Yellow push button	Piano key inputs (Do–Si)	7	1.99 RON	13.93 RON
4×4 Button Matrix Keypad	Menu navigation, mode selection and control input	1	3.99 RON	3.99 RON
LCD 16×2 with I2C interface	Displays current mode, notes and system feedback	1	14.99 RON	14.99 RON
Analog 2-axis joystick module	Experimental pitch / expression control via ADC	1	5.35 RON	5.35 RON
Breadboard	Prototyping platform for wiring components	1	—	—
Jumper wires (M-M, M-F, F-F)	Electrical connections between components	—	—	—

Total hardware cost: ~140 RON

Software

Library	Description	Usage
embassy-stm32	Hardware Abstraction Layer (HAL) for STM32 in Embassy	GPIO (buttons + matrix scan), ADC (joystick), PWM timers (buzzers), I2C (LCD)
embassy-executor	Async task executor for embedded Rust	runs the async #[embassy_executor::main] entry point and the main event loop
embassy-time	Async timers and delays	scan timing for matrix, debounce-ish delays, periodic loop pacing (Timer::after_*)
defmt	Lightweight embedded logging framework	structured logs for button/keypad events, ADC debug, state transitions
defmt-rtt	RTT (Real-Time Transfer) transport for defmt logs	sends logs to the host while debugging (no UART required)
panic-probe	Panic handler for embedded targets	reports panics through the debug probe/RTT
embedded-hal	Traits for embedded drivers (portable interfaces)	used for I2C traits and PWM trait bounds

Detailed Software Design

The firmware is implemented as a single asynchronous control loop using the Embassy framework. All application logic is centralized in the main loop, with functionality divided into clearly defined stages.

The system is divided into:

• Initializing the components
  • LCD: Finding the I2C address, passing the value to the constants
  • Joystick: Calibrating the center of the joystick
  • Buzzers: Creating the "voices" that will handle the notes
  • Hardcoding the frequencies
  • Attributing different functions for each of the buttons on the matrix
  • Defining different modes (FreePlay, SinglePlay, DoublePlay)
• Main loop
  • Default: FreePlay Mode
  • Await button presses that trigger changes to a different mode
    • Intermediary steps e.g. confirmation etc.
• Async functions
  • These will be called when the main loop will switch control

Links

1. Pico Piano
2. Electric Piano