Low-Resource Cross-Domain Singing Voice Synthesis via Reduced Self-Supervised Speech Representations.
Panos KakoulidisNikolaos EllinasGeorgios VamvoukakisMyrsini ChristidouAlexandra VioniGeorgia ManiatiJunkwang OhGunu JhoInchul HwangPirros TsiakoulisAimilios ChalamandarisPublished in: CoRR (2024)
Keyphrases
- cross domain
- text to speech
- multi domain
- domain adaptation
- emotion recognition
- sentiment classification
- transfer learning
- speech quality
- knowledge transfer
- speech recognition
- fundamental frequency
- target domain
- cross domain learning
- speech sounds
- speech signal
- audio visual
- text categorization
- multiple domains
- automatic speech recognition
- audio features
- web image annotation
- multi modal
- e government
- data analysis
- learning algorithm