Pith. sign in

REVIEW 3 major objections 5 minor 42 references

Diffused Responsibility: Analyzing the Energy Consumption of Generative Text-to-Audio Diffusion Models

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Inference energy of text-to-audio diffusion grows linearly with denoising steps, and quality plateaus by 50 steps.

desk verdict Useful first systematic energy benchmark for seven current text-to-audio diffusion models; the step-scaling and batch findings hold up, but the Pareto claim that 150/200 steps are never optimal rests on unreplicated quality metrics and needs variance bars before it becomes recommendation-grade. read the letter →

arxiv 2505.07615 v2 pith:SBIA3PML submitted 2025-05-12 eess.AS cs.AIcs.LGcs.SD

Riccardo Passoni , Francesca Ronchini , Luca Comanducci , Romain Serizel , Fabio Antonacci This is my paper · ORCID
classification eess.AScs.AIcs.LGcs.SD
keywords text-to-audiogenerationdiffusionmodelsinferenceenergydenoisingstepsParetofrontierefficiencyCLAPscoreFréchetAudioDistance
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper measures how much electricity seven text-to-audio diffusion models consume when generating sound, and asks whether the energy buys better audio. It reports that inference energy grows almost perfectly linearly with the number of denoising steps for every model tested, that generating several clips in one batch cuts per-clip energy but with sharply diminishing returns beyond a batch of about five, and that the configurations with the best quality-per-watt trade-offs all use between 10 and 50 steps. Settings of 150 or 200 steps never land on that trade-off frontier, so the extra computation fails to translate into measurably better audio. If these measurements hold, users can cut energy use substantially by choosing low-step settings and a modest batch size, and model choice matters: the cheapest model consistently costs a fraction of the most expensive ones.

What carries the argument

The load-bearing mechanism is the diffusion inference loop itself, whose cost is controlled by the number of denoising steps: each additional step adds roughly constant GPU work, producing the observed linear energy scaling. Energy is counted per generated clip with a GPU-tracking toolkit, while quality is scored by CLAP (semantic alignment between prompt and audio) and Fréchet Audio Distance (distributional similarity to real audio). The Pareto frontier over model, step, and batch configurations is what converts raw energy numbers into an actionable recommendation about which settings dominate others.

What would settle it

Repeat the step-count and Pareto experiments on a second GPU and with total system power (CPU plus GPU plus RAM) instead of GPU-only readings; if 150- or 200-step configurations ever match the FAD/CLAP-per-kWh of 50-step settings, or if the energy-per-step slope is no longer constant for any model, the central claims fail.

Watch

Extended reading notes

Core claim

The paper's central claim is that the inference energy of text-to-audio diffusion models is a predictable, tunable quantity: across AudioLDM, AudioLDM2, Make-an-Audio, Make-an-Audio-2, Stable Audio Open, Tango, and Tango2, the energy measured per generated ten-second clip rises linearly with the number of denoising steps, with each model having its own slope. Batching multiple generations lowers energy per sample, with the largest gain from a batch size of 2 and little benefit past 5. Combining these energy readings with two quality metrics, the authors find that Pareto-optimal configurations—those where no further quality can be gained without more energy, or vice versa—are concentrated at 10 to 50 inference steps; 150- and 200-step configurations never appear on the frontier. The paper presents this as the first inference-focused energy analysis of text-to-audio diffusion models and as evidence that step count, batch size, and model architecture jointly determine the environmental cost of generative audio.

Load-bearing premise

The measurements assume that GPU-only power readings, taken on one NVIDIA A40 while excluding model loading, CPU, RAM, and storage, faithfully represent the inference energy that determines real-world deployment cost.

Editorial extensions

If this is right

  • Users can cut inference energy by roughly the step-count ratio: halving steps from 100 to 50 nearly halves the measured energy for every model.
  • A batch size of 2 to 5 is the efficiency sweet spot; larger batches add little or no per-sample savings, so generating many clips one by one is the most wasteful strategy.
  • Defaulting to 150 or 200 denoising steps is hard to justify on quality grounds: those settings never achieved a Pareto-optimal FAD or CLAP score in this study.
  • Energy-sensitive deployments should prefer the cheapest model architecture, while quality-driven applications can still pick a heavier model as long as steps stay in the 10 to 50 range.
  • The ranking of models on the frontier depends on the dataset, so application domain should guide model choice as much as raw energy cost.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If GPU power is the dominant term, the linear step-energy law suggests that energy budgets for a generation job could be predicted from step count alone, enabling quota-based scheduling; the paper does not test this extension.
  • A natural next experiment is dynamic step scheduling: if quality plateaus by 50 steps, adaptive early stopping during denoising could be benchmarked against the fixed-step frontier.
  • The GPU-only measurement may not transfer to CPU or heterogeneous inference; replicating the Pareto analysis with total system power would show whether the low-step conclusion is hardware-independent.
  • The same Pareto methodology could be applied to autoregressive text-to-audio models, which the paper explicitly leaves for future work.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper presents a comparative empirical study of the energy consumption of seven text-to-audio (TTA) diffusion models at inference time. Using CodeCarbon on a single NVIDIA A40 GPU, the authors measure GPU energy as a function of the number of denoising steps (10–200) and batch size (1–10), and they combine these energy measurements with CLAP and FAD quality metrics on 300-sample subsets of AudioCaps and Clotho to identify Pareto-optimal configurations. The main claims are: (i) inference energy scales approximately linearly with the number of denoising steps across all models; (ii) batching reduces per-sample energy with diminishing returns beyond batch size 5; and (iii) Pareto-optimal quality/energy trade-offs are achieved with low step counts, typically 10–50, while 150- and 200-step configurations never appear on the frontier. The code is publicly released.

Significance. If the findings hold, this is a useful first step toward energy-aware deployment of TTA models. The study covers seven current models, uses repeated energy measurements, and provides public code, which aids reproducibility. The Pareto analysis is a valuable practical contribution because it offers concrete guidance on step-count choices. However, the central Pareto conclusion is currently supported only by point estimates of CLAP/FAD without uncertainty quantification, and the energy measurements are specific to one GPU and one measurement tool. These limitations materially affect the strength of the claims as stated.

major comments (3)
  1. [4. Pareto Frontier Analysis] The statement that "configurations with 150 or 200 diffusion steps never appear on the Pareto frontier" (last sentence of Section 4) is not supported by the reported data. CLAP and FAD are each computed once per configuration on a 300-sample subset, with no confidence intervals, bootstrap replicates, or seeds reported. The Pareto frontier in Fig. 2 is therefore drawn from point estimates; small FAD/CLAP differences (on the order of 0.01–0.02) between adjacent step counts can change frontier membership under resampling. The sanity check described in Section 4 only compares subset vs full-dataset trends for AudioLDM and Stable Audio Open; it does not quantify the variance of the quality metrics or check frontier membership for all models. Please either provide uncertainty estimates (e.g., bootstrap over the 300 samples) and re-derive the frontier, or soften the claim to "in our measurements, no 150/200-step configuration appeared on the frontier."
  2. [3. Experiments and Results] The text states that "Each experiment was repeated 5 times to compute the average and standard deviation of energy consumption," but neither Fig. 1(a), Fig. 1(b), nor Fig. 2 reports any standard deviation, error bars, or numerical values. Without these, the reader cannot assess whether the "clear linear relationship" in Section 3.1 and the "diminishing returns" in Section 3.2 are statistically distinguishable from noise. Please add error bars or confidence intervals, or provide a table of mean±std for all reported configurations.
  3. [2.2 Energy Consumption evaluation] All conclusions rest on CodeCarbon GPU-only energy readings from a single NVIDIA A40, excluding CPU, RAM, model loading, and storage, and no validation of the tracker against hardware power measurements is reported. Since the Pareto frontier in Fig. 2 is determined by absolute kWh values, the identified "Pareto-optimal" step counts may change on other hardware (e.g., consumer GPUs, integrated GPUs, CPU-only inference). The paper should state this as an explicit limitation and, ideally, include a sensitivity analysis or a discussion of how relative model rankings might differ.
minor comments (5)
  1. [Figures 1(a) and 1(b)] The y-axis label of Fig. 1 appears garbled ("/uni00000014/uni00000048/uni00000016" and "/uni00000014/uni00000048/uni00000015"), apparently a PDF rendering issue; please ensure the label reads "Energy at inference [kWh]" on both panels.
  2. [3.1 Does energy consumption scale with inference steps?] The sentence "we generated a separate 10-second audio sample for each inference step using the same prompt" is ambiguous: does this mean one sample per step count per repeat, or a single sample per step? Please clarify how the 5 repeats were allocated.
  3. [1. Introduction] The claim that "only related work on multimodal image generation has begun to address this issue" is inconsistent with the earlier citation of Douwes et al. [9] on the energy of neural audio synthesis models; please narrow the claim to text-to-audio diffusion models or explicitly acknowledge [9] as related audio work.
  4. [4. Pareto Frontier Analysis] Trimming Clotho audio (originally 15–30 s) to 10 s may affect the FAD distribution and the resulting frontier; please discuss or justify this preprocessing choice.
  5. [2.1 TTA models selected] Table 1 reports inference speed without specifying whether this is wall-clock time from a single run or an average; please clarify the measurement procedure.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the study's claims are directly measured externally and are not derived from their own outputs.

full rationale

This is an empirical measurement study, not a derivation. The three central claims—linear growth of inference energy with denoising steps, reduced per-sample energy with batching, and Pareto frontiers over step counts—are obtained by directly measuring GPU energy with CodeCarbon and computing quality scores with external CLAP/FAD metrics. No equation in the paper defines a predicted quantity in terms of the same quantity being inferred, and no fitted parameter is renamed as a prediction. The authors cite prior work, including some papers co-authored by the current authors (e.g., [13], [14], [31], [40]), but these citations are contextual—used for motivation, tool overview, and the GPU-only measurement convention—and are not load-bearing substitutes for the new measurements. The Pareto frontier is a plot of measured values; the finding that 150/200-step configurations are never on the frontier is an empirical observation contingent on single unreplicated quality measurements, which is a statistical robustness concern, not circularity. The sanity check in Section 4 is also an independent comparison of subset versus full-dataset trends, not a circular validation. Consequently, there are no circular steps, and the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claims are empirical measurements; the only load-bearing premises are the accuracy of CodeCarbon, the sufficiency of GPU-only energy, the validity of CLAP/FAD as quality metrics, the representativeness of a single GPU, and the reliability of 300-sample quality estimates. None of these is fitted or invented; they are standard domain assumptions.

assumptions (5)
  • domain assumption CodeCarbon estimates consumed energy accurately from GPU power monitoring
    The paper uses CodeCarbon to measure kWh and reports only GPU energy, citing prior studies; no calibration against a physical wattmeter is reported. Section 2.2.
  • domain assumption GPU energy alone is a sufficient proxy for total inference energy
    CPU and RAM usage are excluded, justified by prior work [5], [31] that GPU dominates; this affects absolute totals. Section 2.2.
  • domain assumption CLAP score and FAD are valid proxies for audio quality
    Quality evaluation relies on CLAP and FAD as standard metrics; these are established in the literature. Section 2.3.
  • domain assumption Energy measured on a single NVIDIA A40 GPU generalizes across deployment hardware
    All experiments run on one A40; the paper acknowledges batch-size effects are hardware-dependent, but the central comparisons assume A40 readings are representative. Sections 3 and 3.2.
  • domain assumption A 300-sample subset reliably represents full-dataset quality metrics
    Pareto analysis uses 300 samples per dataset; a sanity check on full AudioCaps/Clotho with AudioLDM and Stable Audio Open supports consistency. Section 4.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Diffused Responsibility: Analyzing the Energy Consumption of Generative Text-to-Audio Diffusion Models." pith.science (2026). https://pith.science/paper/SBIA3PML

@misc{pith2026250507615,
  author       = {Pith},
  title        = {Pith review of: Diffused Responsibility: Analyzing the Energy Consumption of Generative Text-to-Audio Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SBIA3PML}},
  note         = {Machine review of arXiv:2505.07615}
}
read the original abstract

Text-to-audio models have recently emerged as a powerful technology for generating sound from textual descriptions. However, their high computational demands raise concerns about energy consumption and environmental impact. In this paper, we conduct an analysis of the energy usage of 7 state-of-the-art text-to-audio diffusion-based generative models, evaluating to what extent variations in generation parameters affect energy consumption at inference time. We also aim to identify an optimal balance between audio quality and energy consumption by considering Pareto-optimal solutions across all selected models. Our findings provide insights into the trade-offs between performance and environmental impact, contributing to the development of more efficient generative audio models.

Figures

Figures reproduced from arXiv: 2505.07615 by the authors.

Figure 1
Figure 1. Energy consumption at inference as a function of the number of inference steps (a) and batch size (b). The legend indicates the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Pareto frontier for TTA models. The legend shows the model by color, and the point shape indicates the number of inference steps. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 32 canonical work pages

  1. [1]

    Deep learning for audio signal processing,

    H. Purwins, B. Li, T. Virtanen, J. Schlüter, S.-Y . Chang, and T. Sainath, “Deep learning for audio signal processing,” IEEE Journal of Selected Topics in Signal Processing , 2019

    work page 2019

  2. [2]

    Dong, Generative AI for Music and Audio

    H.-W. Dong, Generative AI for Music and Audio . University of California, San Diego, 2024

    work page 2024

  3. [3]

    Foundation models for music: A survey,

    Y . Ma, A. Øland, A. Ragni, B. M. Del Sette, C. Saitis, C. Donahue, C. Lin, C. Plachouras, E. Benetos, E. Shatri et al., “Foundation models for music: A survey,” arXiv preprint arXiv:2408.14340 , 2024

    arXiv 2024

  4. [4]

    Energy Consumption of Deep Generative Audio Models

    C. Douwes, P. Esling, and J.-P. Briot, “Energy consumption of deep generative audio models,” arXiv preprint arXiv:2107.02621 , 2021

    work page Pith review arXiv 2021

  5. [5]

    Energy consumption trends in sound event detection systems,

    C. Douwes and R. Serizel, “Energy consumption trends in sound event detection systems,” in ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2025

    work page 2025

  6. [6]

    The computational limits of deep learning,

    N. C. Thompson, K. Greenewald, K. Lee, G. F. Manso et al. , “The computational limits of deep learning,” arXiv preprint arXiv:2007.05558 , vol. 10, 2020

    arXiv 2007

  7. [7]

    Green ai,

    R. Schwartz, J. Dodge, N. A. Smith, and O. Etzioni, “Green ai,” Communications of the ACM , 2020

    work page 2020

  8. [8]

    Green mir?: Investigating computational cost of recent music-ai research in ismir,

    A. Holzapfel, A.-K. Kaila, and P. Jääskeläinen, “Green mir?: Investigating computational cost of recent music-ai research in ismir,” in International Society for Music Information Retrieval Conference (ISMIR) , 2024

    work page 2024

Show all 42 references
  1. [9]

    Is quality enoughƒ integrating energy consumption in a large-scale evaluation of neural audio synthesis models,

    C. Douwes, G. Bindi, A. Caillon, P. Esling, and J.-P. Briot, “Is quality enoughƒ integrating energy consumption in a large-scale evaluation of neural audio synthesis models,” in ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)....

    2023

  2. [10]

    Performance above all? energy consumption vs. performance for machine listening, a study on dcase task 4 baseline,

    R. Serizel, S. Cornell, and N. Turpault, “Performance above all? energy consumption vs. performance for machine listening, a study on dcase task 4 baseline,” in ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2023

    2023

  3. [11]

    The energy and carbon footprint of training end-to-end speech recognizers,

    T. Parcollet and M. Ravanelli, “The energy and carbon footprint of training end-to-end speech recognizers,” in Proc. Interspeech 2021, 2021

    2021

  4. [12]

    Sustainability and the internet of sounds: Case studies,

    L. Gabrielli, E. Principi, and L. Turchet, “Sustainability and the internet of sounds: Case studies,” IEEE Transactions on Technology and Society , 2024

    2024

  5. [13]

    Monitoring environmental impact of dcase systems: Why and how?

    C. Douwes, F. Ronchini, and R. Serizel, “Monitoring environmental impact of dcase systems: Why and how?” in Detection and Classification of Acoustic Scene and Events (DCASE) Workshop , 2023

    2023

  6. [14]

    Performance and energy balance: A comprehensive study of state-of-the-art sound event detection systems,

    F. Ronchini and R. Serizel, “Performance and energy balance: A comprehensive study of state-of-the-art sound event detection systems,” in ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2024

    2024

  7. [15]

    Power hungry processing: Watts driving the cost of ai deployment?

    S. Luccioni, Y . Jernite, and E. Strubell, “Power hungry processing: Watts driving the cost of ai deployment?” in The 2024 ACM Conference on Fairness, Accountability, and Transparency, 2024

    2024

  8. [16]

    Diffusion models: A comprehensive survey of methods and applications,

    L. Yang, Z. Zhang, Y . Song, S. Hong, R. Xu, Y . Zhao, W. Zhang, B. Cui, and M.-H. Yang, “Diffusion models: A comprehensive survey of methods and applications,” ACM Computing Surveys , 2023

    2023

  9. [17]

    The environmental impact of deep learning for audio,

    C. Douwes, “The environmental impact of deep learning for audio,” 2023

    2023

  10. [18]

    AudioLDM: Text-to-audio generation with latent diffusion models,

    H. Liu, Z. Chen, Y . Yuan, X. Mei, X. Liu, D. Mandic, W. Wang, and M. D. Plumbley, “AudioLDM: Text-to-audio generation with latent diffusion models,” in Proceedings of the 40th International Conference on Machine Learning , ser. Proceedings of Machine Learning Research, A. Kra...

    2023

  11. [19]

    Clap learning audio concepts from natural language supervision,

    B. Elizalde, S. Deshmukh, M. Al Ismail, and H. Wang, “Clap learning audio concepts from natural language supervision,” in ICASSP 2023- 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023

    2023

  12. [20]

    Audioldm 2: Learning holistic audio generation with self-supervised pretraining,

    H. Liu, Y . Yuan, X. Liu, X. Mei, Q. Kong, Q. Tian, Y . Wang, W. Wang, Y . Wang, and M. D. Plumbley, “Audioldm 2: Learning holistic audio generation with self-supervised pretraining,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , vol. 32, p. 2871–2883, May 2024. [Online]. Av...

    2024

  13. [21]

    Make-an-audio: text-to-audio generation with prompt- enhanced diffusion models,

    R. Huang, J. Huang, D. Yang, Y . Ren, L. liu, M. Li, Z. Ye, J. Liu, X. Yin, and Z. Zhao, “Make-an-audio: text-to-audio generation with prompt- enhanced diffusion models,” in Proceedings of the 40th International Conference on Machine Learning , ser. ICML’23. JMLR.org, 2023

    2023

  14. [22]

    Make-an-audio 2: Temporal-enhanced text-to-audio generation,

    J. Huang, Y . Ren, R. Huang, D. Yang, Z. Ye, C. Zhang, J. Liu, X. Yin, Z. Ma, and Z. Zhao, “Make-an-audio 2: Temporal-enhanced text-to-audio generation,” CoRR, vol. abs/2305.18474, 2023. [Online]. Available: https://doi.org/10.48550/arXiv.2305.18474

    2023 doi arXiv

  15. [23]

    Text-to- audio generation using instruction guided latent diffusion model,

    D. Ghosal, N. Majumder, A. Mehrish, and S. Poria, “Text-to- audio generation using instruction guided latent diffusion model,” in Proceedings of the 31st ACM International Conference on Multimedia, ser. MM ’23. New York, NY , USA: Association for Computing Machinery, 2023, p. ...

    2023

  16. [24]

    The flan collection: Designing data and methods for effective instruction tuning,

    S. Longpre, L. Hou, T. Vu, A. Webson, H. W. Chung, Y . Tay, D. Zhou, Q. V . Le, B. Zoph, J. Wei et al., “The flan collection: Designing data and methods for effective instruction tuning,” in International Conference on Machine Learning. PMLR, 2023

    2023

  17. [25]

    Stable audio open,

    Z. Evans, J. D. Parker, C. Carr, Z. Zukowski, J. Taylor, and J. Pons, “Stable audio open,” in ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2025

    2025

  18. [26]

    Carbontracker: Tracking and predicting the carbon footprint of training deep learning models,

    L. F. W. Anthony, B. Kanding, and R. Selvan, “Carbontracker: Tracking and predicting the carbon footprint of training deep learning models,” arXiv preprint arXiv:2007.03051 , 2020

    2007 arXiv

  19. [27]

    Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters,

    J. Rasley, S. Rajbhandari, O. Ruwase, and Y . He, “Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters,” in Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining , 2020

    2020

  20. [28]

    Codecarbon: estimate and track carbon emissions from machine learning computing,

    V . Schmidt, K. Goyal, A. Joshi, B. Feld, L. Conell, N. Laskaris, D. Blank, J. Wilson, S. Friedler, and S. Luccioni, “Codecarbon: estimate and track carbon emissions from machine learning computing,” Cited on, vol. 20, 2021

    2021

  21. [29]

    Thop: Pytorch-opcounter,

    L. Zhu, “Thop: Pytorch-opcounter,” https://github.com/Lyken17/pytorch- OpCounter, 2019, accessed: 2025-04-24

    2019

  22. [30]

    mlco2/codecarbon: v2.4.1,

    B. Courty, V . Schmidt, S. Luccioni, Goyal-Kamal, MarionCoutarel, B. Feld, J. Lecourt, LiamConnell, A. Saboni, Inimaz, supatomic, M. Léval, L. Blanche, A. Cruveiller, ouminasara, F. Zhao, A. Joshi, A. Bogroff, H. de Lavoreille, N. Laskaris, E. Abati, D. Blank, Z. Wang, A. Cato...

    2024 doi

  23. [31]

    From computation to consumption: Exploring the compute-energy link for training and testing neural networks for sed systems,

    C. Douwes and R. Serizel, “From computation to consumption: Exploring the compute-energy link for training and testing neural networks for sed systems,” arXiv preprint arXiv:2409.05080 , 2024

    2024 arXiv

  24. [32]

    Fr \’echet audio distance: A metric for evaluating music enhancement algorithms,

    K. Kilgour, M. Zuluaga, D. Roblek, and M. Sharifi, “Fr \’echet audio distance: A metric for evaluating music enhancement algorithms,” arXiv preprint arXiv:1812.08466, 2018

    2018 arXiv

  25. [33]

    Audiocaps: Generating captions for audios in the wild,

    C. D. Kim, B. Kim, H. Lee, and G. Kim, “Audiocaps: Generating captions for audios in the wild,” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 2019

    2019

  26. [34]

    Clotho: An audio captioning dataset,

    K. Drossos, S. Lipping, and T. Virtanen, “Clotho: An audio captioning dataset,” in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2020

    2020

  27. [35]

    Audio set: An ontology and human- labeled dataset for audio events,

    J. F. Gemmeke, D. P. Ellis, D. Freedman, A. Jansen, W. Lawrence, R. C. Moore, M. Plakal, and M. Ritter, “Audio set: An ontology and human- labeled dataset for audio events,” in 2017 IEEE international conference on acoustics, speech and signal processing (ICASSP) . IEEE, 2017

    2017

  28. [36]

    Gans trained by a two time-scale update rule converge to a local nash equilibrium,

    M. Heusel, H. Ramsauer, T. Unterthiner, B. Nessler, and S. Hochreiter, “Gans trained by a two time-scale update rule converge to a local nash equilibrium,” Advances in neural information processing systems , 2017

    2017

  29. [37]

    Fr \’echet audio distance: A reference-free metric for evaluating music enhancement algorithms,

    D. Roblek, K. Kilgour, M. Sharifi, and M. Zuluaga, “Fr \’echet audio distance: A reference-free metric for evaluating music enhancement algorithms,” in Proc. Interspeech, 2019

    2019

  30. [38]

    Adapting frechet audio distance for generative music evaluation,

    A. Gui, H. Gamper, S. Braun, and D. Emmanouilidou, “Adapting frechet audio distance for generative music evaluation,” in Proc. IEEE ICASSP 2024, 2024. [Online]. Available: https://arxiv.org/abs/2311.01616

    2024 arXiv

  31. [39]

    Stable audio metrics toolkit,

    S. AI, “Stable audio metrics toolkit,” https://github.com/stability-ai/stable- audio-metrics, 2024, accessed: 2025-05-06

    2024

  32. [40]

    Paguri: a user experience study of creative interaction with text-to-music models,

    F. Ronchini, L. Comanducci, G. Perego, and F. Antonacci, “Paguri: a user experience study of creative interaction with text-to-music models,” arXiv preprint arXiv:2407.04333 , 2024

    2024

  33. [41]

    Sound designer-generative ai interactions: Towards designing creative support tools for professional sound designers,

    P. Kamath, F. Morreale, P. L. Bagaskara, Y . Wei, and S. Nanayakkara, “Sound designer-generative ai interactions: Towards designing creative support tools for professional sound designers,” in Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems , 2024

    2024

  34. [42]

    Freesound datasets: A platform for the creation of open audio datasets

    E. Fonseca, J. Pons, X. Favory, F. Font, D. Bogdanov, A. Ferraro, S. Oramas, A. Porter, and X. Serra, “Freesound datasets: A platform for the creation of open audio datasets.” in ISMIR, 2017

    2017

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.