What This Is, Honestly
Audio Scapes started as a side experiment. A client asked if we could generate custom hold music that didn't sound like a dental office in 1994. We tried Suno, Udio, and a few smaller models. Some of the output was genuinely good. Most of it wasn't. The good ones now run in three production environments.
This page exists because the original WordPress URL was indexed and we want anyone landing here to get a straight answer instead of a 404.
Status: Experimental. We take a few of these a quarter. We're not trying to scale it into a service line.
What We Actually Do
Generate ambient audio for spaces where music plays in the background but nobody is listening on purpose. Hold lines, lobby spaces, brand videos, internal training modules. The brief is usually some combination of mood, tempo, instrumentation, and "doesn't sound like a stock library."
We use a mix of models depending on the brief — different tools handle different genres differently. The output gets evaluated against the brief, the best 2-3 candidates get iterated on, and you pick the final track. Licensing is sorted before we ship anything, in writing.
Who This Is For
A specific kind of buyer:
- Boutique hospitality (a coffee shop, a yoga studio, a small hotel) that wants a sonic identity but can't justify a $15K composer commission.
- A SaaS company replacing the default Twilio hold music on its support line.
- A B2B brand producing video assets and tired of the same five Epidemic Sound tracks everyone else uses.
Who This Is Not For
- You need licensed music with vocals. Use a music licensing platform (Musicbed, Epidemic Sound, Artlist). Cheaper, faster, legally cleaner.
- You need a jingle, a podcast theme, or anything memorable. Hire a human composer. AI is not there yet for hook-driven music.
- You need film scoring or any synced-to-picture work. Different craft entirely.
What It Costs
| Engagement | What you get | Investment | |------------|--------------|------------| | Single track | One 2–4 minute looping track, brief-to-delivery in about a week, two revision rounds | $400 | | Track set | Three to five thematically linked tracks for a brand library, four revision rounds | $1,200–$2,000 | | Custom hold-line system | Tracks plus the IVR audio handoff, plus the Twilio configuration if you want it managed | Quoted per engagement |
These are real prices, not anchor pricing. Smaller engagements aren't worth the overhead for either party.
What You Get and What You Don't
You get:
- Final tracks in WAV and MP3
- Documented license for your specific use case (commercial, internal, etc.)
- A short written note on the model used, the prompt, and the iteration history — so you or another vendor can pick up where we left off
You don't get:
- Exclusivity. Other clients can generate from similar prompts and end up with similar moods. We don't guarantee uniqueness, because we can't.
- Performance rights for venues that require ASCAP/BMI licensing. AI-generated music is in legal flux on this; we won't pretend otherwise.
- A roadmap of forever-ongoing tracks. This is a finite engagement.
The Honest Caveats
AI-generated audio is improving fast. The track we generate today may sound dated in 12 months as the models improve. We tell every client this. Some don't care because the use case is a hold line nobody listens to closely. Some care because the use case is a brand identity, and we work with them to define a refresh cadence.
Copyright law around AI-generated audio is unsettled. We follow the current state of the law — clean prompts, no training-data attacks, no impersonation of named artists. If the legal landscape shifts in a way that affects what we shipped, we'll tell you.
How to Start
Send us a brief: where the audio lives, what mood, what tempo, what instruments you like, what to avoid. A two-paragraph email is enough. We'll either say yes with a quote, or we'll tell you why this isn't the right fit and point you somewhere better.
Hank Groman | hank@lastapple.com | 949-529-9017