Generative Spoken Dialogue Language Modeling - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Transactions of the Association for Computational Linguistics Year : 2023

Generative Spoken Dialogue Language Modeling

Abstract

We introduce dGSLM, the first "textless" model able to generate audio samples of naturalistic spoken dialogues. It uses recent work on unsupervised spoken unit discovery coupled with a dual-tower transformer architecture with cross-attention trained on 2000 hours of two-channel raw conversational audio (Fisher dataset) without any text or labels. We show that our model is able to generate speech, laughter and other paralinguistic signals in the two channels simultaneously and reproduces more naturalistic and fluid turn taking compared to a text-based cascaded model.
Fichier principal
Vignette du fichier
2023.tacl-1.15.pdf (785.59 Ko) Télécharger le fichier
Origin : Publisher files allowed on an open archive
Licence : CC BY - Attribution

Dates and versions

hal-03985368 , version 1 (13-02-2023)
hal-03985368 , version 2 (23-01-2024)

Licence

Attribution

Identifiers

Cite

Tu Anh Nguyen, Eugene Kharitonov, Jade Copet, Yossi Adi, Wei-Ning Hsu, et al.. Generative Spoken Dialogue Language Modeling. Transactions of the Association for Computational Linguistics, 2023, 11, pp.250-266. ⟨10.1162/tacl_a_00545⟩. ⟨hal-03985368v2⟩
130 View
50 Download

Altmetric

Share

Gmail Facebook X LinkedIn More