BERTIVITS: The Posterior Encoder Fusion of Pre-Trained Models and Residual Skip Connections for End-to-End Speech Synthesis

Enhancing the naturalness and rhythmicity of generated audio in end-to-end speech synthesis is crucial. The current state-of-the-art (SOTA) model, VITS, utilizes a conditional variational autoencoder architecture. However, it faces challenges, such as limited robustness, due to training solely on te...

Descripción completa

Detalles Bibliográficos

Autores principales:	Zirui Wang, Minqi Song, Dongbo Zhou
Formato:	Artigo
Lenguaje:	Inglês
Publicado:	MDPI AG 2024-06-01
Colección:	Applied Sciences
Materias:	pre-trained model text to speech neural TTS speech synthesis end-to-end model
Acceso en línea:	https://www.mdpi.com/2076-3417/14/12/5060
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!