This is the demo page for the paper “AI TrackMate: Finally, Someone Who’ll Give Your Music More Than Just ’Sounds Great!’”.

Abstract

The rise of “bedroom producers” has democratized music creation, but challenges producers to objectively evaluate their work. We present AI TrackMate, an LLM-based music chatbot designed to provide constructive feedback on music productions. Unlike previous approaches, AI TrackMate leverages the inherent musical knowledge of LLMs and focuses on production-oriented feedback. Our framework consists of a Music Analysis Module, an LLM Readable Music Report, and Production-Oriented Feedback Instruction for LLM. This plug-and-play, training-free system is compatible with various LLMs and adaptable to future advancements. We demonstrate AI TrackMate through an interactive web interface and conduct a pilot study with a music producer. By bridging AI capabilities with the needs of independent producers, AI TrackMate offers round-the-clock, nuanced feedback, potentially transforming the creative process and skill development in music production.

System

System Overview

The system comprises three layers:

  1. User Interface for audio upload, query input, and feedback reception
  2. Data Processing for handling raw audio and text
  3. AI Analysis, featuring a Music Analysis Module that transforms raw audio into LLM-readable reports, and an LLM that processes these reports along with user queries. Guided by music production-oriented feedback instructions, the LLM generates insights comparable to those of a music producer.

Example

AI Producer Dialogue History

Yi-Lin Jiang: rogerylc@gmail.com
Chia-Ho Hsiung: chiahohsiung@gmail.com
Yen-Tung Yeh: r12942179@ntu.edu.tw
Lu-Rong Chen: warconvict@gmail.com
Bo-Yu Chen: bernie40916@gmail.com