Augmentation Invariant Discrete Representation for Generative Spoken Language Modeling - CNRS - Centre national de la recherche scientifique Access content directly
Conference Papers Year : 2023

Augmentation Invariant Discrete Representation for Generative Spoken Language Modeling

Représentation discrète invariante de l'augmentation pour la modélisation générative de la parole

Abstract

Generative Spoken Language Modeling research focuses on optimizing speech Language Models (LMs) using raw audio recordings without accessing any textual supervision. Such speech LMs usually operate over discrete units obtained from quantizing internal representations of self-supervised models. Although such units show impressive modeling results, their robustness capabilities have not been extensively investigated. This work focuses on improving the invariance of discrete input representations to non-spoken augmentations for generative spoken language modeling. First, we formally define how to measure the robustness of such representations to various signal variations that do not alter the spoken information (e.g., time-stretch). Next, we empirically demonstrate how current state-of-the-art representation models lack robustness to such variations. To overcome this, we propose an effective and efficient method to learn invariant discrete speech representation for generative spoken language modeling. The proposed approach is based on applying a set of signal transformations to the speech signal and optimizing the model using an iterative pseudolabeling scheme. Our method significantly improves over the evaluated baselines when considering encoding and modeling metrics. We additionally evaluate our method on the speechto-speech translation task, considering Spanish-English and French-English translations, and show the proposed approach outperforms the evaluated baselines.
Fichier principal
Vignette du fichier
2209.15483.pdf (454.16 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04208443 , version 1 (21-09-2023)

Identifiers

Cite

Itai Gat, Felix Kreuk, Tu Anh Nguyen, Ann Lee, Jade Copet, et al.. Augmentation Invariant Discrete Representation for Generative Spoken Language Modeling. 20th International Conference on Spoken Language Translation (IWSLT 2023), Jul 2023, Toronto, Canada. pp.465-477, ⟨10.18653/v1/2023.iwslt-1.46⟩. ⟨hal-04208443⟩
88 View
81 Download

Altmetric

Share

Gmail Facebook X LinkedIn More