A probabilistic approach to selecting units for speech synthesis based on acoustic similarity.
Anjana BabuK. Raghava KrishnanAnil Kumar SaoHema A. MurthyPublished in: NCC (2014)
Keyphrases
- speech synthesis
- prosodic features
- speech recognition
- text to speech
- similarity measure
- vocal tract
- probabilistic model
- generative model
- similarity function
- posterior probability
- bayesian networks
- similarity measurement
- hidden markov models
- distance measure
- distance metric
- pattern recognition
- semantic similarity
- uncertain data
- similarity metric
- feature selection
- speech corpus
- similarity scores
- structural similarity
- selection algorithm
- multilayer perceptron
- conditional probabilities
- euclidean distance
- language model
- feature space