mlsanityData to 7 Oct 2026

Adapting Frechet Audio Distance for Generative Music Evaluation

arXiv 2311.01616 ↗First posted 2 Nov 2023eess.AScited by 92

Not on the map. Why?

Cited by 29 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The growing popularity of generative music models underlines the need for perceptually relevant, objective music quality metrics. The Frechet Audio Distance (FAD) is commonly used for this purpose even though its correlation with perceptual quality is understudied. We show that FAD performance may be hampered by sample size bias, poor choice of audio embeddings, or the use of biased or low-quality reference sets. We propose reducing sample size bias by extrapolating scores towards an infinite sample size. Through comparisons with MusicCaps labels and a listening test we identify audio embeddings and music reference sets that yield FAD scores well-correlated with acoustic and musical quality. Our results suggest that per-song FAD can be useful to identify outlier samples and predict perceptual quality for a range of music sets and generative models. Finally, we release a toolkit that allows adapting FAD for generative music evaluation.

Key facts. Cited by 92 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 Nov 2023.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 32; 2025: 30; 2026: 29.

92 papers.

Sort:

By area:

  1. 2024DITTO: Diffusion Inference-Time T-Optimization for Music Generationcited by 73
  2. 2024Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversioncited by 58
  3. 2025ACE-Step: A Step Towards Music Generation Foundation Modelcited by 54
  4. 2024Masked Generative Video-to-Audio Transformers with Enhanced Synchronicitycited by 28
  5. 2024DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music Generationcited by 26
  6. 2026Self-Supervised Flow Matching for Scalable Multi-Modal Synthesiscited by 25
  7. 2025KAD: No More FAD! An Effective and Efficient Evaluation Metric for Audio Generationcited by 24
  8. 2024Beyond FVD: Enhanced Evaluation Metrics for Video Generation Qualitycited by 21
  9. 2025Aligning Text-to-Music Evaluation with Human Preferencescited by 20
  10. 2024Presto! Distilling Steps and Layers for Accelerating Music Generationcited by 16
  11. 2025FlowDec: A flow-based full-band general audio codec with high perceptual qualitycited by 15
  12. 2025Benchmarking Music Generation Models and Metrics via Human Preference Studiescited by 15
38 more
  1. 2024COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representationscited by 14
  2. 2024A Diffusion-Based Generative Equalizer for Music Restorationcited by 12
  3. 2024Frechet Music Distance: A Metric For Generative Symbolic Music Evaluationcited by 12
  4. 2024MusiConGen: Rhythm and Chord Control for Transformer-Based Text-to-Music Generationcited by 11
  5. 2024Unsupervised Blind Joint Dereverberation and Room Acoustics Estimation with Diffusion Modelscited by 11
  6. 2024Towards Assessing Data Replication in Music Generation with Music Similarity Metrics on Raw Audiocited by 10
  7. 2025Survey on the Evaluation of Generative Models in Musiccited by 10
  8. 2024Latent Diffusion Bridges for Unsupervised Musical Audio Timbre Transfercited by 9
  9. 2024Diversity-Rewarded CFG Distillationcited by 9
  10. 2024Naturalistic Music Decoding from EEG Data via Latent Diffusion Modelscited by 8
  11. 2025MIDI-VALLE: Improving Expressive Piano Performance Synthesis Through Neural Codec Language Modellingcited by 8
  12. 2025Music Arena: Live Evaluation for Text-to-Musiccited by 8
  13. 2023CMMD: Contrastive Multi-Modal Diffusion for Video-Audio Conditional Modelingcited by 7
  14. 2024Quality-aware Masked Diffusion Transformer for Enhanced Music Generationcited by 6
  15. 2024Challenge on Sound Scene Synthesis: Evaluating Text-to-Audio Generationcited by 6
  16. 2025Unified Cross-modal Translation of Score Images, Symbolic Music, and Performance Audiocited by 6
  17. 2025AudioMorphix: Training-free audio editing with diffusion probabilistic modelscited by 6
  18. 2026The Design Space of Tri-Modal Masked Diffusion Modelscited by 6
  19. 2026MuQ-Eval: An Open-Source Per-Sample Quality Metric for AI Music Generation Evaluationcited by 6
  20. 2024Nested Music Transformer: Sequentially Decoding Compound Tokens in Symbolic Music and Audio Generationcited by 5
  21. 2026Stable Audio 3cited by 5
  22. 2024FastSAG: Towards Fast Non-Autoregressive Singing Accompaniment Generationcited by 4
  23. 2024An Independence-promoting Loss for Music Generation with Language Modelscited by 4
  24. 2024Unsupervised Composable Representations for Audiocited by 4
  25. 2025DRAGON: Distributional Rewards Optimize Diffusion Generative Modelscited by 4
  26. 2025MGE-LDM: Joint Latent Diffusion for Simultaneous Music Generation and Source Extractioncited by 4
  27. 2025Re-Bottleneck: Latent Re-Structuring for Neural Audio Autoencoderscited by 4
  28. 2025SoundReactor: Frame-level Online Video-to-Audio Generationcited by 4
  29. 2025Steering Autoregressive Music Generation with Recursive Feature Machinescited by 4
  30. 2024MambaFoley: Foley Sound Generation using Selective State-Space Modelscited by 3
  31. 2024Semi-Supervised Cognitive State Classification from Speech with Multi-View Pseudo-Labelingcited by 3
  32. 2025A Multimodal Symphony: Integrating Taste and Sound through Generative AIcited by 3
  33. 2025Controllable Video-to-Music Generation with Multiple Time-Varying Conditionscited by 3
  34. 2026TuneJury: An Open Metric for Improving Music Generation Preference Alignmentcited by 3
  35. 2024WaveTransfer: A Flexible End-to-end Multi-instrument Timbre Transfer with Diffusioncited by 2
  36. 2025Diffused Responsibility: Analyzing the Energy Consumption of Generative Text-to-Audio Diffusion Modelscited by 2
  37. 2025Recomposer: Event-roll-guided generative audio editingcited by 2
  38. 2025Mitigating data replication in text-to-audio generative diffusion models through anti-memorization guidancecited by 2

50 most cited of 92 All 92 (JSON)