Tijana Nosek
Papers
1
Total Citations
11
H-Index
1
About
Tijana Nosek is a researcher in speech synthesis and voice technology, with a focus on creating flexible, data-efficient systems for generating natural-sounding speech. Her key research areas include neural network-based speech synthesis, speaker adaptation, and style transfer in spoken language. Nosek’s most notable contribution is her work on speaker and style embedding, which enables neural networks to synthesize speech in a specific voice or speaking style using only a small amount of training data. This approach, detailed in her highly cited 2020 paper, maps discrete speaker and style variables into continuous embeddings, allowing for more personalized and expressive synthetic voices without requiring extensive recordings. Her method has significant implications for accessibility tools, virtual assistants, and entertainment, where rapid adaptation to new speakers or styles is crucial. With 11 citations on this foundational paper, Nosek’s work is gaining recognition for its practical efficiency and technical elegance. Her research bridges the gap between high-quality synthesis and real-world constraints, making her a promising voice in the field of speech technology.
Research Focus
Key Achievements
Top Papers
- 1