Normal-to-shouted speech spectral mapping for speaker recognition under vocal effort mismatch.
Ana Ramírez LópezRahim SaeidiLauri JuvelaPaavo AlkuPublished in: ICASSP (2017)
Keyphrases
- speaker recognition
- speech signal
- spectral analysis
- speaker identification
- speech recognition
- speaker verification
- gaussian mixture model
- vector quantization
- automatic speech recognition
- probabilistic neural network
- acoustic features
- hidden markov models
- emotion recognition
- mel frequency cepstral coefficients
- spectral features
- non stationary
- noisy environments
- broadcast news
- audio visual
- information fusion
- noisy images
- maximum likelihood
- similarity measure
- feature selection