Text-independent voice conversion using speaker model alignment method from non-parallel speech.
Peng SongYun JinWenming ZhengLi ZhaoPublished in: INTERSPEECH (2014)
Keyphrases
- statistical model
- theoretical analysis
- mathematical model
- evaluation method
- probabilistic model
- prior knowledge
- cost function
- similarity measure
- modeling method
- linear model
- objective function
- pairwise
- input data
- parameter estimation
- em algorithm
- text to speech
- prediction model
- optimization method
- energy function
- autoregressive
- speech recognition
- feature set