Speech style conversion based on the statistics of vowel spectrograms and nonlinear frequency mapping.
Toru TakahashiHideki BannoToshio IrinoHideki KawaharaPublished in: EUSIPCO (2006)
Keyphrases
- speech signal
- fundamental frequency
- speech recognition
- vocal tract
- formant frequencies
- speech synthesis
- automatic speech recognition
- spectral density
- noisy environments
- speaker identification
- prosodic features
- linear prediction
- speech recognition systems
- text to speech
- speaker independent
- hidden markov models
- non stationary
- data sets
- low frequency
- speaker verification
- complementarity problems
- pattern recognition
- nonlinear models
- high frequency
- dimensionality reduction
- information retrieval