APEX: Large-Scale Multi-Task Aesthetic-Informed Popularity Prediction for AI-Generated Music

APEX is the first large-scale multi-task learning framework for jointly predicting popularity and aesthetic quality of AI-generated music from audio alone. It is trained on over 211k AI-generated songs (~10k hours of audio) from Suno and Udio, leveraging MERT-v1-95M audio embeddings.

---

What does APEX predict?

Given any audio file, APEX predicts 7 scores:

Popularity:

Score	Range	Description
score_streams	0–100	Predicted streaming engagement score
score_likes	0–100	Predicted likes engagement score

Aesthetic Quality (from SongEval):

Score	Range	Description
coherence	1–5	Structural and harmonic coherence
musicality	1–5	Overall musical quality
memorability	1–5	How memorable the song is
clarity	1–5	Clarity of production and mix
naturalness	1–5	Naturalness of the generated audio

---

Architecture

---

Usage

Installation

pip install torch transformers soundfile torchaudio

Inference

from transformers import AutoModel

model   = AutoModel.from_pretrained("amaai-lab/apex", trust_remote_code=True)
results = model.predict("my_song.mp3", save_json="results.json")

print(results["score_streams"])  # popularity score 0-100
print(results["score_likes"])    # popularity score 0-100
print(results["coherence"])      # aesthetic score 1-5