Eachlabs | AI Workflows for app builders
mureka-describe-song

MUREKA

Mureka Describe Song is a music analysis model that generates descriptive insights about a song’s structure, style, and characteristics.

Avg Run Time: 15.000s

Model Slug: mureka-describe-song

Playground

Input

Enter a URL or choose a file from your computer.

Output

Example Result

Preview and download your result.

{
"output":{
"description":"This is a high-energy pop-R&B track with a polished and modern production style. The song features a clear and smooth male lead vocal, supported by layered backing vocals that add harmonic richness. The rhythm section is driven by a crisp electronic drum beat and a groovy synth bass line, creating an infectious and danceable groove. The arrangement is filled out with lush synth pads and bright synth keys, contributing to a full and vibrant soundscape. The overall mood is overwhelmingly positive, motivational, and uplifting, making it ideal for content that aims to inspire and energize its audience."
"genres":[
0:"Pop"
1:"R&B"
2:"Soul Pop"
]
"instrument":[
0:"Lead Vocal (Male, Clear, Smooth)"
1:"Backing Vocals"
2:"Electronic Drum Kit"
3:"Synth Bass"
4:"Synth Pad"
5:"Synth Keys"
]
"status":"COMPLETED"
"tags":[
0:"upbeat"
1:"motivational"
2:"inspirational"
3:"positive"
4:"energetic"
5:"driving"
6:"polished"
7:"contemporary"
8:"feel-good"
9:"uplifting"
]
"trace_id":"e230024f4e4bcfb68f762a07cff98e13"
}
}
Each execution costs $0.1000. With $1 you can run this model about 10 times.

API & SDK

Create a Prediction

Send a POST request to create a new prediction. This will return a prediction ID that you'll use to check the result. The request should include your model inputs and API key.

Get Prediction Result

Poll the prediction endpoint with the prediction ID until the result is ready. The API uses long-polling, so you'll need to repeatedly check until you receive a success status.

Readme

Table of Contents
Overview
Technical Specifications
Key Considerations
Tips & Tricks
Capabilities
What Can I Use It For?
Things to Be Aware Of
Limitations

Overview

mureka-describe-song — Text-to-Audio AI Model

Developed by Mureka as part of the mureka family, mureka-describe-song is a specialized music analysis model that generates detailed descriptive insights into a song’s structure, style, vocals, and musical characteristics from audio input. Unlike standard text-to-music generators, it reverses the process by dissecting uploaded tracks to reveal AI-driven breakdowns of elements like BPM, time signatures, melody motifs, and arrangement logic—empowering creators with transparent analysis for refinement and inspiration. This makes mureka-describe-song ideal for musicians seeking AI music analysis tools to understand and iterate on their compositions quickly.

Part of Mureka's advanced ecosystem featuring MusiCoT technology, mureka-describe-song leverages chain-of-thought reasoning to map song structures coherently, providing outputs that highlight verse-chorus transitions, vocal timbres, and instrumentation choices.

Technical Specifications

What Sets mureka-describe-song Apart

mureka-describe-song stands out in the AI music analysis landscape through its integration of Mureka's MusiCoT (Music Chain-of-Thought) framework, which plans and describes song elements logically before output. This enables precise breakdowns of complex tracks, such as identifying drum machines, bass arpeggios, and pad selections, giving users actionable insights into AI decision-making that generic analyzers overlook.

It supports multi-modal inputs like reference tracks, hummed melodies, or full audio uploads in formats including MP3, WAV, and FLAC, with analysis completing in minutes for tracks up to 240 seconds. This versatility sets it apart for Mureka AI music tools users analyzing professional-grade content.

  • Transparent Structure Mapping: Uses CoT reasoning to outline verse, chorus, and bridge logic, helping producers spot weaknesses in arrangement completeness—unlike basic spectrum analyzers.
  • Vocal and Style Insights: Describes AI vocal timbres, multi-language support (10+ languages), and cloning compatibility, enabling voice branding consistency in revisions.
  • Segment-Level Detail: Breaks down musicality, BPM contrasts, and emotional builds, supporting edits without full regenerations for efficient workflows.

Key Considerations

  • Use detailed prompts specifying genre, mood, tempo, vocal style, and energy for best musical control and adherence to intent
  • Include structural markers like [intro], [verse], [chorus] in lyrics to guide song organization and improve coherence
  • Balance quality settings during generation to access advanced features without compromising output
  • Increase character limits (up to 1000 for prompts, 3000 for lyrics) for complex inputs in newer versions like V7.5
  • Avoid vague prompts; specify vocal gender (male/female) or instrumental-only for targeted results
  • Test with reference audio uploads for style guidance, but ensure compatibility with model versions

Tips & Tricks

How to Use mureka-describe-song on Eachlabs

Access mureka-describe-song seamlessly through Eachlabs' Playground for instant analysis by uploading audio files (MP3, WAV, up to 240s) or text/lyrics prompts, or integrate via API/SDK with parameters like track references and style filters for structured JSON insights on song elements. Outputs deliver high-fidelity descriptions of structure, vocals, and MusiCoT reasoning, ready for commercial use in minutes.

---

Capabilities

  • Generates complete songs from lyrics with automatic melody, harmony, vocals, and arrangements
  • Produces expressive, natural-sounding vocals with emotional dynamics and professional instrumentation
  • Supports controllable generation via text prompts for genre, mood, tempo, structure, and reference audio
  • Handles long-form music up to 6 minutes with maintained coherence and diversity
  • Offers specialized modes: short-music for videos and per-section style control (e.g., intro vs. chorus)
  • Provides high subjective quality in motifs, vocals, structure, and resonance per benchmarks

What Can I Use It For?

Use Cases for mureka-describe-song

Music producers can upload a reference track to mureka-describe-song for a full breakdown of its melodic motifs and vocal textures, then use the insights to extend or remix sections—saving hours on manual stem separation for film scoring projects.

Content creators analyzing short video audio via AI song description tools benefit from detailed style reports, like "catchy synth-pop chorus with 128 BPM and layered pads," to match vibes across social media clips without music theory expertise.

Developers integrating mureka-describe-song API into apps for automated music analysis can process user-uploaded songs, receiving JSON outputs on structure and instrumentation to power recommendation engines or auto-editing features—for gaming soundtracks or ad jingles.

Songwriters feed a hummed melody or lyrics-backed track into the model with a prompt like "Describe the structure, mood, and vocal style of this indie rock demo," gaining feedback on chorus contrast and emotional arcs to refine before full production.

Things to Be Aware Of

  • Newer models like V8 excel in emotional vocals and harmony but require balanced generation settings
  • API updates frequently add features like stems, voice cloning, and higher limits, so check version defaults (e.g., V7.5)
  • Users report strong performance in subjective quality over competitors, especially in arrangement and resonance
  • Custom vocals via uploads enhance personalization but need MP3 samples and management
  • Resource needs scale with song length; short modes optimize for quick video backgrounds
  • Positive feedback highlights natural flow and control, with consistent outperformance in benchmarks

Limitations

  • Primarily a generation model, not focused on song description or voice classification as named; no direct "describe-song" functionality found
  • Relies heavily on quality of input prompts and lyrics; vague inputs may yield less coherent results
  • Parameter counts and exact training data undisclosed, limiting reproducibility details

Pricing

Pricing Detail

This model runs at a cost of $0.10 per execution.

Pricing Type: Fixed

The cost remains the same regardless of which model you use or how long it runs. There are no variables affecting the price. It is a set, fixed amount per run, as the name suggests. This makes budgeting simple and predictable because you pay the same fee every time you execute the model.