VoiceID on the fly: A Speaker Recognition System that Learns from Scratch - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

VoiceID on the fly: A Speaker Recognition System that Learns from Scratch

Résumé

We proposed a novel AI framework to conduct real-time multi-speaker recognition without any prior registration or pretraining by learning the speaker identification on the fly. We considered the practical problem of online learning with episodically revealed rewards and introduced a solution based on semi-supervised and self-supervised learning methods in a web-based application at https://www.baihan.nyc/viz/VoiceID/.
Fichier principal
Vignette du fichier
InterSpeech_VoiceID_updated.pdf (575.07 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02939812 , version 1 (15-09-2020)

Identifiants

  • HAL Id : hal-02939812 , version 1

Citer

Baihan Lin, Xinxin Zhang. VoiceID on the fly: A Speaker Recognition System that Learns from Scratch. INTERSPEECH, Oct 2020, Shanghai, China. ⟨hal-02939812⟩
149 Consultations
127 Téléchargements

Partager

Gmail Facebook X LinkedIn More