END-TO-END SPEECH RECOGNITION FROM FEDERATED ACOUSTIC MODELS - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

END-TO-END SPEECH RECOGNITION FROM FEDERATED ACOUSTIC MODELS

Résumé

Training Automatic Speech Recognition (ASR) models under federated learning (FL) settings has attracted a lot of attention recently. However, the FL scenarios often presented in the literature are artificial and fail to capture the complexity of real FL systems. In this paper, we construct a challenging and realistic ASR federated experimental setup consisting of clients with heterogeneous data distributions using the French and Italian sets of the CommonVoice dataset, a large heterogeneous dataset containing thousands of different speakers, acoustic environments and noises. We present the first empirical study on attention-based sequence-to-sequence Endto-End (E2E) ASR model with three aggregation weighting strategies-standard FedAvg, loss-based aggregation and a novel word error rate (WER)-based aggregation, compared in two realistic FL scenarios: cross-silo with 10 clients and cross-device with 2K and 4K clients. Our analysis on E2E ASR from heterogeneous and realistic federated acoustic models provides the foundations for future research and development of realistic FL-based ASR applications.
Fichier principal
Vignette du fichier
2104.14297.pdf (447.18 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03601224 , version 1 (08-03-2022)

Identifiants

  • HAL Id : hal-03601224 , version 1

Citer

Yan Gao, Titouan Parcollet, Salah Zaiem, Javier Fernandez-Marques, Pedro de Gusmao, et al.. END-TO-END SPEECH RECOGNITION FROM FEDERATED ACOUSTIC MODELS. The International Conference on Acoustics, Speech, & Signal Processing (ICASSP), May 2022, Singapour, Singapore. ⟨hal-03601224⟩

Collections

UNIV-AVIGNON LIA
42 Consultations
61 Téléchargements

Partager

Gmail Facebook X LinkedIn More