fspecii/ace-step-ui

ACE-Step UI is a web-based front-end for generating AI music locally on your own computer, built to work alongside the ACE-Step 1.5 open-source music generation model. The idea is that instead of paying a monthly subscription to services like Suno or Udio, you run the AI model on your own hardware and use this interface to control it, keeping your music private and removing any generation limits. The interface is designed to look and feel like a music streaming app, with a library view, a bottom playback bar with waveform display, dark and light modes, and playlist management. From the generation side, you can describe the style of music you want using genre tags and a lyrics editor, and the model produces full songs up to about four minutes long, including vocals. There are also controls for instrumental-only output, batch generation of multiple variations at once, and a seed control so you can reproduce a specific result later. Beyond basic generation, the tool includes a reference audio feature where you can supply an existing audio file and have the model match its style. A repainting option lets you regenerate specific sections of a track rather than starting over. Stem extraction powered by Demucs can separate a finished track into its vocal, drum, bass, and other components. There is also a built-in audio editor for trimming and applying effects, and a video generator that pairs music with background footage from Pexels. On the technical side, the front-end is built with React and TypeScript, the back-end runs on Express with a local SQLite database, and communication with the ACE-Step model happens through its Gradio API. Requirements include an NVIDIA GPU with at least 4 GB of video memory, Node.js 18 or higher, Python 3.10 or above, and FFmpeg for audio processing. Installation can be done through Pinokio, a one-click app launcher that handles all dependencies automatically, or manually via shell scripts on Windows, macOS, and Linux. The full README is longer than what was shown.