Pith. sign in

REVIEW 5 cited by

Correlation of Fr\'echet Audio Distance With Human Perception of Environmental Audio Is Embedding Dependant

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.17508 v1 pith:4KQ52OF3 submitted 2024-03-26 cs.SD eess.AS

Modan Tailleur , Junwon Lee , Mathieu Lagrange , Keunwoo Choi , Laurie M. Heller , Keisuke Imoto , Yuki Okamoto This is my paper · ORCID
classification cs.SDeess.AS
keywords audiocorrelationechetembeddingembeddingsenvironmentalperceptualdistance
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

This paper explores whether considering alternative domain-specific embeddings to calculate the Fr\'echet Audio Distance (FAD) metric can help the FAD to correlate better with perceptual ratings of environmental sounds. We used embeddings from VGGish, PANNs, MS-CLAP, L-CLAP, and MERT, which are tailored for either music or environmental sound evaluation. The FAD scores were calculated for sounds from the DCASE 2023 Task 7 dataset. Using perceptual data from the same task, we find that PANNs-WGM-LogMel produces the best correlation between FAD scores and perceptual ratings of both audio quality and perceived fit with a Spearman correlation higher than 0.5. We also find that music-specific embeddings resulted in significantly lower results. Interestingly, VGGish, the embedding used for the original Fr\'echet calculation, yielded a correlation below 0.1. These results underscore the critical importance of the choice of embedding for the FAD metric design.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Smooth-Foley: Creating Continuous Sound for Video-to-Audio Generation Under Semantic Guidance

    cs.SD 2024-12 conditional novelty 6.0 of 10

    Smooth-Foley uses frame-level visual features and label-guided temporal conditions to generate continuous, synchronized audio for videos with moving or ambiguous sound sources.

  2. CoDiCodec: Unifying Continuous and Discrete Compressed Representations of Audio

    cs.SD 2025-09 conditional novelty 5.0 of 10

    CoDiCodec unifies continuous and discrete audio compression in one consistency-trained autoencoder, using FSQ-dropout to serve both continuous ~11 Hz embeddings and 2.38 kbps discrete tokens.

  3. The Name-Free Gap: Policy-Aware Stylistic Control in Music Generation

    cs.SD 2025-08 conditional novelty 5.0 of 10

    Word-based style descriptors generated by an LLM can shift MusicGen outputs toward a target artist's sound almost as much as using the artist's name, defining a name-free gap.

  4. Continuous Autoregressive Models with Noise Augmentation Avoid Error Accumulation

    cs.LG 2024-11 conditional novelty 5.0 of 10

    Injecting random noise into input embeddings during training lets purely autoregressive models generate continuous audio embeddings without quality degradation over long sequences.

  5. Frechet Music Distance: A Metric For Generative Symbolic Music Evaluation

    cs.SD 2024-12 conditional novelty 4.0 of 10

    FMD applies the Frechet distance to CLaMP music embeddings to quantify distributional similarity between generated and reference symbolic music.

Pith tools