CHATR the Corpus; a 20-year-old archive of Concatenative Speech Synthesis.
Nick CampbellPublished in: LREC (2016)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- vocal tract
- prosodic features
- manually annotated
- speech corpus
- metadata
- test set
- digital archives
- hidden markov models
- spontaneous speech
- computer vision
- sentence level
- linguistic features
- statistical machine translation
- supervised machine learning
- coreference resolution
- image processing