A user study on the influence of mobile device class, synthesis method, data rate and lexicon on speech synthesis quality.
Michael PucherPeter FröhlichPublished in: INTERSPEECH (2005)
Keyphrases
- synthetic data
- input data
- test data
- data sets
- correlation analysis
- user input
- statistical methods
- data quality
- noisy data
- data collection
- prior knowledge
- missing data
- data points
- statistical analysis
- speech synthesis
- data sources
- data analysis
- clustering method
- end users
- user satisfaction
- high quality
- training data
- similarity measure
- high dimensional data
- missing values
- hidden markov models
- training set