56,951 songs indexed

56,000+ Songs. AI-Interpreted.
One API.

Music-critic-quality prose interpretations, 15 structured attributes, and 768-dimensional embeddings for every song in the catalog. Built for recommendation engines, playlist tools, and music intelligence products.

GET /api/v1/songs/search?q=bohemian+rhapsody
{
  "title": "Bohemian Rhapsody",
  "artist": "Queen",
  "genre_primary": "Progressive Rock",
  "mood_primary": "Epic",
  "energy": 8,
  "tempo_feel": "variable",
  "interpretation": "A six-minute odyssey that
    defies every convention of popular music...",
  "embedding": [0.0231, -0.0087, ...] // 768 dims
}

What You Get

Three layers of music intelligence, from human-readable prose to machine-ready vectors.

Prose Interpretations
Music-critic-quality analysis of every song

Avg 1,272 characters per song. Covers musical structure, lyrical themes, production techniques, cultural context, and emotional impact. Written by AI trained on music journalism.

ProEnterprise
15 Structured Attributes
Machine-readable musical DNA
genresubgenremoodenergytempoemotional_arcvocal_styleproductionvalencedanceabilitytextureacousticnesseracultural_originlistening_context
ProEnterprise
768-dim Embeddings
Semantic vectors for similarity and ML

OpenAI text-embedding-3-small vectors computed from structured interpretations. Use for similarity search, clustering, recommendation engines, and as ML features.

Enterprise

See the Data

Real API response for a song detail request. Every song gets this level of analysis.

GET/api/v1/songs/search?q=butter&page_size=1
200 OK
{
  "data": [{
    "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "title": "Butter",
    "artist": "BTS",
    "genre_primary": "Pop",
    "subgenre": "Dance Pop",
    "mood_primary": "Joyful",
    "energy": 8,
    "tempo_feel": "medium",
    "era_feel": "2020s",
    "sonic_texture": "polished, bright, layered synths",
    "interpretation": "A masterclass in pop songwriting
      precision. BTS delivers an irresistibly smooth
      English-language track built on a buttery disco-funk
      bassline and retro synth stabs...",
    "structured_text": "This Pop / Dance Pop song has a
      Joyful mood with energy level 8/10...",
    "embedding": [0.0142, -0.0298, 0.0511, ...]
  }],
  "meta": {
    "page": 1, "page_size": 1, "total": 3
  }
}

Start in 3 Lines

Simple REST API. No SDK required. Works with any language.

curl -s "https://dashboard-ruddy-beta.vercel.app/api/v1/songs/search?q=dynamite" \
  -H "Authorization: Bearer YOUR_API_KEY" | jq .

Built For

Music data infrastructure for products that understand music.

Recommendation Engines

Use embeddings for semantic similarity search. Find songs that feel alike, not just share tags.

Playlist Curation

Auto-generate playlists using mood, energy, and emotional arc attributes. Build coherent listening journeys.

Music Journalism

Jumpstart editorial content with AI interpretations. Research song context, production analysis, and cultural impact.

Sync Licensing

Match songs to scenes using structured attributes. Filter by mood, energy, tempo, and production style.

The Catalog

Growing weekly. K-Pop, Pop, Rock, Hip-Hop, R&B, Jazz, Electronic, Classical, and more.

56,951
Songs
12,053
Artists
596
Genres
528
Moods
15
Attributes / Song

Simple Pricing

Start free. Scale when you need to.

Free
Get started exploring the catalog
$0
  • 100 requests / day
  • Basic song fields
  • No interpretations
  • No embeddings
Get Started
Pro
Popular
For products that need music intelligence
$49 / mo
  • 10,000 requests / day
  • All basic fields
  • Prose interpretations
  • Structured text
  • No embeddings
Subscribe
Enterprise
Full access with vectors and export
$299 / mo
  • 100,000 requests / day
  • Everything in Pro
  • 768-dim embeddings
  • Similarity search
  • Bulk NDJSON export
Subscribe