Power-up sound effects
A power-up sound has to be ascending and in a major key to read as a reward. The prompt words for that are few and easy to learn.
Listen: 11 samples from the power-up collection
Every clip shows the exact prompt that produced it. The render script reads its prompts from the same file as this page. Copy one, change a word and generate your own.
Triumphant power chant (vocal)
via audiocpp-ace-steptriumphant orchestral-hybrid power-up theme, uplifting choir vocals, bright brass and strings, ascending major key
Double jump
via audiocpp-stable-audio-3-sfxvideo game double jump power-up, short bright upward pitch sweep with an airy tail, dry
Weapon upgrade
via audiocpp-stable-audio-3-sfxweapon upgrade sound, metallic mechanical lock-in followed by a rising energy charge and a bright snap
Grow
via audiocpp-stable-audio-3-sfxvideo game grow power-up, comedic ascending pitch slide with a soft inflating whoosh, bright and dry
Berserk mode
via audiocpp-stable-audio-3-sfxberserk mode activation, deep distorted energy surge rising into a growling sustained low drone, aggressive
Time slow
via audiocpp-stable-audio-3-sfxtime slowing power-up, everything pitching down and stretching into a low warped drone with a soft flutter
Magnet active
via audiocpp-stable-audio-3-sfxmagnetic pull power-up, humming electromagnetic drone with a rising pitched shimmer, clean and synthetic
Fire enchant
via audiocpp-stable-audio-3-sfxfire enchantment activation, whoosh of igniting flame wrapping around metal with a crackling tail
Mega charge
via audiocpp-stable-audio-3-sfxmega power-up charge, long rising synthetic energy build with an accelerating pulse, ending in a bright burst
Stat boost
via audiocpp-stable-audio-3-sfxstat boost confirmation, three clean ascending bell tones with a warm sustain, dry
Overdrive loop
via audiocpp-stable-audio-3-sfxpower-up overdrive active loop, pulsing synthetic energy hum with a bright shimmering top layer, seamless loop
Prompts that work
These models respond to physical description. Name the material, the action and the space.
Classic power-up
chiptune power-up jingle, five-note ascending major arpeggio, square wave, no decay tail, 500 ms
"No decay tail" keeps each note punchy and equal in volume, so the five-note arpeggio reads as one event and doesn't trail off.
Shield activate
shield activation sound, rising synth pad with a soft shimmer, 600 ms, warm low-mid body, subtle reverb
"Warm low-mid body" signals protection. A shield sound needs weight and coverage; a thin bright lead sounds like speed.
Speed boost
speed boost whoosh-synth hybrid, fast rising pitch sweep with a bright tail, 400 ms, energetic, dry
"Whoosh" with "synth" blends an air-movement sound with a musical rise. The blend reads as velocity, which a plain arpeggio does not.
Invincibility
invincibility power-up fanfare, rapid ascending square wave run across two octaves, sparkly, 700 ms, dry
Spanning "two octaves" makes this sound bigger and rarer than an ordinary five-note power-up. A wider range signals a bigger effect.
Health restore
health restore chime, warm rising sine wave with a gentle bell overtone, 500 ms, soft reverb tail
A sine wave with only a "gentle" bell overtone keeps the sound soothing, which separates a healing effect from an attack or speed buff.
Making power-up sounds that fit
Match the timbre to what the power-up does
All power-ups move upward in pitch, but the timbre should describe what the buff does. Speed wants brightness and a fast sweep. Shields and armor want low-mid warmth and weight. Healing wants a soft sine-based tone with no harsh overtones. Offensive buffs (invincibility, damage boosts) can be the brightest and loudest of the set. Write the effect into the prompt as well as the word "power-up".
Scale duration and range to rarity
A common pickup should be quick (three to five notes, under 500 ms) so it doesn't interrupt gameplay when it happens often. A rare buff like invincibility can run longer and span a wider pitch range, because the player hears it less often and the extra half second marks it as special.
Keep the whole power-up family on one waveform
If your game has five different buffs, generate all of them from the same base waveform (square wave for a retro game, synth pad for a modern one) and vary only the pitch range, speed and overtone description between prompts. The player then hears "something good happened" before working out which buff it was.
Save the sung chant for the rarest buff
The triumphant chant above is the odd one out: ACE-Step's `lyrics` field gives it a sung line. Use that for the one power-up in your game that should feel like an event (a legendary drop, an ultimate unlock), and avoid it for anything the player hears often. A sung hook that repeats every few minutes gets tiresome.
Why generate locally
The models run on your machine, which changes the cost, the privacy and what you can automate.
Offline after download
Weights are fetched once. After that, generation, the library and the API all work with the network off. Everything lives in one SQLite file and one folder of WAVs under ~/.soundbox, and nothing is uploaded.
Your agent can use it
Soundbox runs an MCP server in the same process, with tools for generating and listing sounds. Generated files come back as absolute paths, so an agent can copy them into a project.
Licences stated up front
Every model's terms are shown before you download it, and non-commercial weights need an explicit confirmation. Soundbox is MIT. The licence of the model that made a sound governs what you may do with it.
claude mcp add --transport http soundbox http://127.0.0.1:17494/mcpcurl -X POST http://127.0.0.1:17494/api/generate \
-H 'Content-Type: application/json' \
-d '{"model":"audiocpp-stable-audio-3-sfx","prompt":"8-bit coin pickup blip","duration_s":1}'Frequently asked questions
- Can I use these sounds in a commercial video or game?
- It depends on the model that made them. Soundbox shows each model's licence before you download its weights. The Stable Audio 3 models use the Stability AI Community License: commercial use is free below USD $1M annual revenue, and above that Stability requires an enterprise licence. ACE-Step 1.5 is MIT. MiniMax's models allow commercial use under a revenue cap and require an attribution notice. Some voice and stem-separation models are non-commercial only. Soundbox itself is MIT, but the audio is governed by the licence of the model that made it. Read that licence before you ship; this summary is not legal advice.
- Does this run offline?
- Yes, once the weights are downloaded. Soundbox runs the models locally and stores everything in one SQLite file and one folder of WAVs under ~/.soundbox. There is no account, API key or upload step. Only the initial model download needs the network.
- What hardware do I need?
- An NVIDIA GPU with 8 GB or more. CPU-only is supported, but each generation then takes minutes. That is fine for a few short clips and slow for a large batch.
- Can my coding agent generate these for me?
- Yes. While Soundbox is running it serves MCP at http://127.0.0.1:17494/mcp, with tools for generating and listing sounds. An agent can generate a sound as part of the task it is already doing and gets back an absolute file path to use in your project.
- How is this different from downloading a sample pack?
- A sample pack's power-up sounds were tuned for someone else's buff system, and often only the filename tells a shield from a speed boost. Generating lets you write the effect into the prompt (warmth for a shield, a bright sweep for speed), so each power-up is distinguishable by ear and the family shares a waveform.