Contrastive self-supervised speaker embedding with sequential disentanglement
Contrastive self-supervised learning has been widely used in speaker embedding to
address the labeling challenge. Contrastive speaker embedding assumes that the contrast …
address the labeling challenge. Contrastive speaker embedding assumes that the contrast …
CCSRD: Content-Centric Speech Representation Disentanglement Learning for End-to-End Speech Translation
Deep neural networks have demonstrated their capacity in extracting features from speech
inputs. However, these features may include non-linguistic speech factors such as timbre …
inputs. However, these features may include non-linguistic speech factors such as timbre …
[PDF][PDF] Joint Language and Speaker Classification in Naturalistic Bilingual Adult-Toddler Interactions
Bilingual children at a young age can benefit from exposure to dual language, impacting
their language and literacy development. Speech technology can aid in developing tools to …
their language and literacy development. Speech technology can aid in developing tools to …
Estimating the completeness of discrete speech units
Representing speech with discrete units has been widely used in speech codec and speech
generation. However, there are several unverified claims about self-supervised discrete …
generation. However, there are several unverified claims about self-supervised discrete …
VoxGenesis: Unsupervised Discovery of Latent Speaker Manifold for Speech Synthesis
Achieving nuanced and accurate emulation of human voice has been a longstanding goal in
artificial intelligence. Although significant progress has been made in recent years, the …
artificial intelligence. Although significant progress has been made in recent years, the …