Speech bandwidth extension using Gaussian mixture model-based estimation of the highband mel spectrum.
Hannu PulakkaUlpu RemesKalle J. PalomäkiMikko KurimoPaavo AlkuPublished in: ICASSP (2011)
Keyphrases
- speech recognition
- accurate estimation
- maximum likelihood
- estimation algorithm
- maximum likelihood estimation
- correlation function
- audio visual
- gaussian distribution
- speech signal
- emotion recognition
- speech synthesis
- continuous valued
- mixture model
- density estimation
- automatic speech recognition
- wireless networks
- linear prediction
- spoken language
- text to speech