Výsledky vyhledávání - "van Staden, Lisa"

Report

A comparison of self-supervised speech representations as input features for unsupervised acoustic word embeddings

Many speech processing tasks involve measuring the acoustic similarity between speech segments. Acoustic word embeddings (AWE) allow for efficient comparisons by mapping speech segments of arbitrary duration to fixed-dimensional vectors. For zero-res

Externí odkaz: http://arxiv.org/abs/2012.07387

Zobrazit plný text záznamu

Report

Unsupervised acoustic unit discovery for speech synthesis using discrete latent-variable neural networks

Autor: Eloff, Ryan, Nortje, André, van Niekerk, Benjamin, Govender, Avashna, Nortje, Leanne, Pretorius, Arnu, van Biljon, Elan, van der Westhuizen, Ewald, van Staden, Lisa, Kamper, Herman

For our submission to the ZeroSpeech 2019 challenge, we apply discrete latent-variable neural networks to unlabelled speech and use the discovered units for speech synthesis. Unsupervised discrete subword modelling could be useful for studies of phon

Externí odkaz: http://arxiv.org/abs/1904.07556

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání