Exemplar-based sparse representation of timbre and prosody for voice conversion.
Huaiping MingDong-Yan HuangLei XieShaofei ZhangMinghui DongHaizhou LiPublished in: ICASSP (2016)
Keyphrases
- sparse representation
- text to speech
- synthesized speech
- speech synthesis
- dictionary learning
- sparse coding
- prosodic features
- face recognition
- image classification
- reconstruction error
- compressive sensing
- image patches
- image representation
- high dimensional data
- compressed sensing
- sparse reconstruction
- random projections
- joint optimization
- dimensionality reduction
- signal processing
- piecewise smooth
- regularized least squares
- acoustic features
- speaker verification
- negative matrix factorization
- data analysis
- neural network