A Comprehensive Analysis of Tokenization and Self-Supervised Learning in End-to-End Automatic Speech Recognition applied on French Language - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

A Comprehensive Analysis of Tokenization and Self-Supervised Learning in End-to-End Automatic Speech Recognition applied on French Language

Résumé

The performance of end-to-end automatic speech recognition (ASR) systems enables their increasing integration into numerous applications. While there are various benefits to such speech-to-text systems, the choice of hyperparameters and models plays a crucial role in their performance. Typically, these choices are determined by considering only the character (CER) and/or word error rate (WER) metrics. However, it has been shown in several studies that these metrics are largely incomplete and fail to adequately describe the downstream application of automatic transcripts. In this paper, we conduct a qualitative study on the French language that investigates the impact of subword tokenization algorithms and self-supervised learning models from different linguistic and acoustic perspectives, using a comprehensive set of evaluation metrics.
Fichier principal
Vignette du fichier
Roux__Eusipco_2024____Analysis_of_end_to_end-1.pdf (137.52 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04584931 , version 1 (23-05-2024)

Identifiants

  • HAL Id : hal-04584931 , version 1

Citer

Thibault Bañeras-Roux, Mickael Rouvier, Jane Wottawa, Richard Dufour. A Comprehensive Analysis of Tokenization and Self-Supervised Learning in End-to-End Automatic Speech Recognition applied on French Language. 32th European Signal Processing Conference (EUSIPCO), 2024, Lyon, France. ⟨hal-04584931⟩
45 Consultations
44 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More