HMM-based speech synthesis of live sports commentaries: integration of a two-layer prosody annotation.
Benjamin PicartSandrine BrognauxThomas DrugmanPublished in: SSW (2013)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- hidden markov models
- prosodic features
- vocal tract
- semantic annotation
- real time
- active learning
- speech corpus
- image retrieval
- machine learning
- application layer
- digital libraries
- multi layer
- noisy environments
- automatic annotation
- pattern recognition
- metadata
- data mining
- middle layer
- databases