Speech segment selection for concatenative synthesis based on prosody-aligned distance measure.
Chih-Chung KuoChi-Shiang KuoPublished in: ICASSP (2002)
Keyphrases
- speech signal
- distance measure
- speech recognition
- speech synthesis
- euclidean distance
- similarity measure
- noisy environments
- distance metric
- hausdorff distance
- texture features
- vector space
- dynamic time warping
- distance function
- nearest neighbor classifier
- prosodic features
- hamming distance
- dissimilarity measure
- edit distance
- kullback leibler divergence
- cosine similarity
- text to speech
- kullback leibler
- data points
- fisher criterion
- pattern recognition