Performance above all ? energy consumption vs. performance for machine listening, a study on dcase task 4 baseline - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

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

Abstract

In machine listening there is a tendency to resort to models with a growing number of parameters raising thus concerns about the practical viability of these due to their energy consumption. Reporting energy consumption of the models could be a first step to raise awareness on this matter. Yet, estimating the energy consumption across different conditions (hyper-parameters, GPU types etc.) poses some challenges in terms of biases and fairness of the comparison between different models and works. In this paper we perform an extensive study using the DCASE task 4 baseline system and monitor energy consumption and training time for different GPU types and batch sizes. The goal is to identify which aspects can have an impact on the estimation of the energy consumption and should be normalized for a fair comparison across systems. Additionally, we propose an analysis of the relationship between the energy consumption and the sound event detection performance that calls into question our current way to evaluate systems.
Fichier principal
Vignette du fichier
ICASSP_2023_task_4.pdf (115.39 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03850797 , version 1 (14-11-2022)

Licence

Identifiers

Cite

Romain Serizel, Samuele Cornell, Nicolas Turpault. Performance above all ? energy consumption vs. performance for machine listening, a study on dcase task 4 baseline. ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun 2023, Rhodes Island, France. pp.1-5, ⟨10.1109/ICASSP49357.2023.10095938⟩. ⟨hal-03850797⟩
132 View
158 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More