timoncool/ACE-Step-Studio

Portable AI music generator — full songs with vocals, covers, music videos. One-click install, 100% offline, NVIDIA GPU.

What it solves

ACE-Step Studio provides a fully local, private, and free alternative to cloud-based AI music services like Suno. It allows users to generate complete songs—including vocals, lyrics, and instrumentation—without requiring an internet connection, subscriptions, or API keys.

How it works

The studio is built on the ACE-Step 1.5 XL (a 4B parameter DiT music generation model). It combines this core audio engine with a local Language Model (LM) for generating lyrics and style descriptions. The system is designed as a portable Windows application that bundles its own Python and Node.js environments, making it accessible via a one-click installer. Users can choose between different model versions (Turbo for speed, SFT for quality) and can optionally integrate cloud APIs (OpenRouter and Pollinations.ai) for enhanced lyrics or album art generation.

Who it’s for

Musicians, content creators, and AI enthusiasts who want to generate high-quality AI music locally on their own hardware (specifically NVIDIA GPUs with 12GB+ VRAM) to ensure privacy and avoid recurring costs.

Highlights

  • Full Song Generation: Creates songs up to 8 minutes long with vocals in any language or genre.
  • Advanced Audio Control: Includes "Cover" mode to transform styles, "Repaint" mode for regenerating specific song sections, and reference audio guidance.
  • Integrated Video Studio: A built-in visualizer and karaoke lyrics generator that exports videos in various aspect ratios (16:9, 9:16, 1:1) using FFmpeg acceleration.
  • Comprehensive Toolset: Features stem extraction via Demucs, a built-in audio editor (AudioMass), and model management tools for merging models or baking LoRAs.
  • Local-First Design: Operates 100% offline by default, with optional opt-in cloud features for those who prefer them.

Related

  • Project
    ServeurpersoCom/acestep.cpp

    A local AI music generation server powered by GGML that turns text descriptions into stereo 48kHz audio using the ACE-Step 1.5 models.

  • Project
    ace-step/ACE-Step-1.5

    ACE‑Step 1.5 is an open‑source music‑generation foundation model that combines a language‑model planner with a Diffusion‑Transformer audio decoder. It can create full songs (10 s‑10 min) in under 10 seconds on a RTX 3090, runs with <4 GB VRAM for base models, and supports LoRA fine‑tuning from a handful of tracks. The repo provides Gradio and REST interfaces, launch scripts for all major OSes, multiple model sizes (including a 4 B XL version), and extensive control over lyrics, style, tempo, and instrumentation.

  • Project
    techjarves/Uncensored-Local-Studio

    A zero-configuration, offline AI studio that integrates Stable Diffusion, LLMs, Whisper, and Kokoro TTS into a single private desktop interface with hardware acceleration.

  • Project
    debpalash/VoiceStudio

    A local-first audio workstation for voice cloning, dubbing, and transcription, providing a unified interface for 27+ AI audio engines.

  • Project
    jamiepine/voicebox

    An open-source, local-first AI voice studio for voice cloning, speech generation, and system-wide dictation, featuring an MCP server for AI agent integration.