Samples

Samples is an audio generator that creates music and sound effects locally using Stable Audio 3.

Audio Studio MCP generates full instrumental tracks and sound effects locally on an Apple Silicon Mac using the Stable Audio 3 medium model. Users input text prompts and seeds to produce unedited audio files without manual mixing. The system handles music and background tracks, detailed sound effects, crossfaded audio loops, and audio editing functions like track extension and style transformation. It is designed for creators who want local audio generation. Supported features include generating music from single sentences, describing sound effects by source, action, material, and space, looping audio tails into heads, and editing existing tracks.

Samples at a glance

Best for: Software engineer
Runs on: macOS
Pricing: Free
Released: October 2026
Category: AI audio tool
Website: kanaworksai-hue.github.io

Key features

Generate instrumental tracks
Create sound effects
Loop audio tails
Extend tracks
Re-imagine tracks in other styles

Use it for

Generating background music for videos
Creating game sound effects
Producing custom instrumentals
Looping audio tracks

Alternatives to Samples

audio.cpp — Runs modern audio models locally with high-performance, native C++ execution.
MLX Audio — It enables fast audio model inference directly on Apple M-series chips.
Blokkily — Pattern-first DAW with synchronized tracker, piano roll, step views, and MCP control
Mubert — Generates royalty-free music instantly using artificial intelligence and human samples.
Macvoice — Runs locally on Apple Silicon in under 2 GB with fast response times.
FluidAudio — Runs local audio AI models on Apple Neural Engine.
dbb1 radio — Generates streaming radio stations directly from descriptive scenes
AuK — It handles zero-shot TTS, editing, and enhancement via natural language.
runs in your browser — It runs locally in your web browser using WebGPU.
AudioCraft — Single codebase for generative music, sound effects, and audio compression.
Free ASR — Zero-dependency real-time streaming speech-to-text with no account login required.
Super Bass Fully Agentic Dexed — Generates synthesizer patches using natural language input.

All Samples alternatives