The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units - Inria - Institut national de recherche en sciences et technologies du numérique
Conference Papers Year : 2020

The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units

Abstract

We present the Zero Resource Speech Challenge 2020, which aims at learning speech representations from raw audio signals without any labels. It combines the data sets and metrics from two previous benchmarks (2017 and 2019) and features two tasks which tap into two levels of speech representation. The first task is to discover low bit-rate subword representations that optimize the quality of speech synthesis; the second one is to discover word-like units from unsegmented raw speech. We present the results of the twenty submitted models and discuss the implications of the main findings for unsupervised speech learning.
Fichier principal
Vignette du fichier
interspeech_2020___ZR2020_Summary_paper__Ewan_.pdf (1014.88 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02962224 , version 1 (09-10-2020)

Identifiers

  • HAL Id : hal-02962224 , version 1

Cite

Ewan Dunbar, Julien Karadayi, Mathieu Bernard, Xuan-Nga Cao, Robin Algayres, et al.. The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units. Interspeech 2020 - Conference of the International Speech Communication Association, Oct 2020, Shangai / Virtual, China. ⟨hal-02962224⟩
172 View
139 Download

Share

More