Higher Education Category Submission

333 – PIZZ

PIZZ is an AI-powered platform that transforms music education by solving three core problems: access barriers, learning difficulties, and lack of feedback. It combines real-time visual feedback, comprehensive performance reports, and gamified learning to make music practice smarter, more inclusive, and engaging.

At its core, PIZZ uses fast audio analysis (FFT + CNN) and real-time MIDI mapping to detect pitch/rhythm errors instantly. A custom AI model then generates color-coded feedback and personalized retraining cycles. Users receive comprehensive performance reports with quantified scoring and future analysis on technique and expressiveness using vision and audio models (e.g., RNN, MediaPipe).

PIZZ also features a streak-based motivational system, a social learning hub for score sharing and collaboration, and an NFT-based recognition system. Blockchain ensures authenticity and ownership of achievements.

The platform is designed for high impact in education: it reduces reliance on expensive private lessons, provides 24/7 feedback, and bridges gaps caused by teacher shortages. It emphasizes inclusivity, learning equity, and long-term engagement.

Music learners often face three challenges: lack of access to quality teachers, difficulty in self-correction, and poor motivation due to delayed or absent feedback. These barriers are worsened by economic inequality and geographic limitations.

The idea for PIZZ was inspired by observing how musicians struggle to identify and correct their mistakes without real-time guidance. Our hypothesis is that combining real-time AI feedback with gamified motivation and community features can dramatically improve learning outcomes and retention.

PIZZ addresses fundamental frustrations in music learning with instant error detection, personalized correction modes, and performance analytics. It replaces one-size-fits-all learning with adaptive, data-driven education. By offering a free/low-cost solution accessible on any device, PIZZ democratizes quality music education and empowers learners worldwide.

PIZZ leverages a combination of audio signal processing, machine learning, and modern web technologies to deliver an effective, scalable solution for music education. At the core of its real-time feedback system is the Fast Fourier Transform (FFT), which processes live audio input to extract pitch and rhythm information. This data is then analyzed using convolutional neural networks (CNNs) to detect deviations from the expected notes and timing. The frontend interface, built with React.js, provides users with intuitive visual feedback using color-coded indicators and timing bars that help learners instantly identify and correct mistakes.

The backend infrastructure, developed using Node.js and Python, supports the core AI analysis and handles data storage and user interactions. Firebase and MySQL are used to manage user profiles, practice streaks, and performance history, while cloud services such as AWS S3 are employed for storing recordings and score content. The platform also includes a practice correction system, which allows users to set personalized retraining rules—automatically looping or restarting sections when errors are detected.

To validate market demand, we plan to conduct pilot testing with music students, educators, and institutions, collecting both usage metrics and qualitative feedback to refine the user experience and identify high-impact features. To measure the system’s effectiveness and accuracy, we will collaborate with experienced musicians and music educators who will test the platform using real performances. These test cases will be analyzed both by PIZZ’s AI and by professional judges, allowing us to compare the AI’s scoring and feedback with expert human evaluations. This will help us calibrate the system’s sensitivity and refine the models to align with professional standards.

As we are applying under Stream 1, this section is not applicable.

PIZZ introduces an innovative and creative approach to solving long-standing problems in music education by combining real-time artificial intelligence, gamification, and blockchain technology in a unified learning platform. Traditionally, music learners rely on delayed feedback from instructors, which limits their ability to self-correct and progress independently. PIZZ disrupts this model by offering instant, AI-powered feedback that analyzes pitch accuracy, rhythm consistency, and dynamic expression in real time. This immediacy empowers learners to identify and fix mistakes as they happen, fostering faster and more effective skill development.

What sets PIZZ apart is its creative integration of multiple cutting-edge technologies into the user experience. The use of color-coded visual feedback and timing bars makes abstract musical concepts tangible and intuitive, especially for beginners. The platform’s customizable “Practice Correction Mode” is another inventive feature that simulates a responsive teacher by looping or restarting sections based on user-defined learning goals. This adaptability ensures that each learner receives a personalized and engaging experience.

Beyond individual learning, PIZZ fosters community and motivation through social features such as score sharing, collaborative composition tools, and a dynamic leaderboard. Its NFT-based achievement system is a novel application of blockchain in education, allowing learners to earn verifiable digital badges and certificates tied to their performance. This not only adds a sense of ownership but also introduces a decentralized way to recognize and showcase musical accomplishments.

Altogether, PIZZ’s innovation lies not just in its use of advanced technology, but in the thoughtful way these tools are designed to enhance learning, accessibility, and motivation. By transforming music education into an interactive, data-driven, and socially connected experience, PIZZ redefines how music can be learned, practiced, and shared in the digital age.

To ensure that PIZZ can scale effectively with increasing user demand, we plan to adopt database and infrastructure strategies that prioritize responsiveness and flexibility. As the platform will handle large volumes of user data—such as audio recordings, performance reports, and practice logs—we intend to use Firebase and MySQL for scalable data management. These technologies will allow us to dynamically expand storage and maintain fast response times even as user numbers grow. To address potential bottlenecks, we will explore techniques such as data indexing, caching, and partitioning to optimize query performance. Media files and user-generated content will be stored on cloud services like AWS S3, which offer scalable and cost-effective storage solutions.

The system will be built using a modular architecture, allowing different components—such as AI analysis, frontend interaction, and gamification features—to scale independently. We plan to containerize backend services and support load balancing so that server resources can be allocated dynamically during peak usage. Our AI models will be optimized for efficient inference using techniques like batching and quantization to support real-time feedback at scale.

From a sustainability perspective, we aim to minimize environmental impact by relying on cloud infrastructure that supports energy-efficient operations. PIZZ will be designed to run smoothly on low-end devices and web browsers, ensuring accessibility without requiring heavy hardware or excessive power consumption.

To foster long-term user engagement, we will incorporate adaptive learning paths, gamified motivation systems, and community features that evolve based on user feedback. We also plan to regularly update content and expand the platform to include new instruments, skill levels, and learning styles. This future-focused approach will ensure that PIZZ remains adaptable, inclusive, and sustainable as both a product and an educational ecosystem.

PIZZ aims to address key social issues in music education by making high-quality learning tools accessible to individuals regardless of socioeconomic background, geographic location, or access to traditional instruction. Many aspiring musicians, especially in underserved or rural communities, lack access to experienced teachers or structured feedback. PIZZ will bridge this gap by providing AI-powered, real-time feedback and personalized practice tools that learners can access from any device with an internet connection. This will enable individuals to improve their musical skills without needing costly lessons or physical infrastructure.

Our solution will also align with broader social goals such as equity and inclusion. The platform will be designed to support diverse learning needs, including customizable difficulty levels, visual feedback for different cognitive learning styles, and multilingual support. We plan to make the core features of PIZZ freely available to ensure that economic barriers do not prevent participation. Furthermore, the social features—such as score sharing, community collaboration, and peer feedback—will help foster a sense of belonging and support among users, especially those who may feel isolated in their musical journey.

To measure social impact, we will track metrics such as user demographics, accessibility adoption rates in low-resource settings, performance improvement over time, and community engagement levels. We also plan to gather qualitative feedback from educators, students, and organizations to assess how the platform is being used and where it can be improved. Regular updates and community input will guide the evolution of PIZZ, ensuring it remains relevant and responsive to user needs. Ultimately, we aim to empower learners, democratize access to music education, and contribute to a more inclusive and supportive global learning environment.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.