Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.samplevault.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Enrichment is the pass that assigns labels — genres, tags, creators, sample packs — onto each individual sample by matching them against the facets available in your library. It’s always free, runs automatically as needed, and you almost never have to trigger it by hand. Enrichment is independent from indexing, which is a separate AI pass that builds the directory-level catalog of allowed facets. The two are often discussed together because most users index a directory and then let enrichment fill samples in afterwards, but enrichment can also run on its own — for example, after you add or rename a tag, or after a sample’s metadata changes.

​
Enrichment vs. Indexing

EnrichmentIndexing
What it doesAssigns existing facets (genres, tags, creators, sample packs) to individual samplesGenerates the catalog of facets available in a directory
CostFreeCounts against Library coverage, per indexed sample
When it runsAutomatically, as needed (after indexing, after edits, after new samples appear)When you choose to index a directory
Triggered byThe app, almost always automaticallyRight-click → Index Directory, or the Add Directory dialog
For local audio processing — BPM, key, spectral features, and sound characteristics — see the Analyzed stage in Adding & Managing Samples. That’s a separate, fully local, deterministic step.

​
How Enrichment Runs

Enrichment is automatic. You don’t normally need to think about it:
  • After you index a directory, enrichment runs across its samples to match the new facets.
  • When new samples are added to an already-indexed directory, enrichment picks them up.
  • When you rename or delete a facet library-wide, enrichment reconciles affected samples.
If you ever want to re-run it manually — for instance, after disabling a label in Enrichment Labels — right-click a directory in the File Browser and choose Enrich Samples. This is rarely necessary. Enrichment also produces two embeddings per sample as it runs: a classification embedding that powers Similar Samples, and a rich-description embedding that powers Prompt Search. These appear in the Library status panel as the Classified and Embedded stages.

​
Indexing a Directory

To create the facet catalog that enrichment matches against, right-click any directory in the File Browser and select Index Directory. The AI scans the folder’s contents and generates labels from filenames, folder structure, and audio content. The Add Directory dialog shows how many samples a folder will draw from your Library coverage before you commit.

​
Managing Enrichment Labels

Control which labels are allowed to be assigned to your samples in Settings > Processing > Enrichment Labels. From here you can manage four label categories:
  • Genres — AI-detected genre classifications
  • Tags — Descriptive labels (e.g., “dark”, “punchy”, “ambient”)
  • Creators — Detected sample pack creators or artists
  • Sample Packs — Detected sample pack names
Toggle individual labels on or off to control what gets assigned. Disabled labels won’t appear on any samples, even if indexing detected them.
If labels are getting assigned that don’t match your library, disable them in Enrichment Labels rather than manually editing each sample. Enrichment will reconcile your library on the next pass.

​
Cost Summary

  • Enrichment is always free.
  • Indexing draws from your plan’s Library coverage pool, measured per indexed sample. If a batch fails, every sample in that batch is refunded in full.
  • Demo Room users get a one-time grant of 5,000 indexed samples. Home Studio includes 50,000 samples per month; Professional Suite includes 200,000.
See Plans & Usage for full details on how the coverage pool works, including monthly rollover and add-on top-ups.