System Performance as a Function of Calibration Methods, Sample Size and Sampling Variability in Likelihood Ratio-Based Forensic Voice Comparison.
Bruce Xiao WangVincent HughesPublished in: Interspeech (2021)
Keyphrases
- sample size
- likelihood ratio
- hypothesis testing
- progressive sampling
- statistical tests
- random sampling
- model selection
- hypothesis test
- covariance matrix
- confidence intervals
- small sample
- upper bound
- small sample size
- generalization error
- statistical hypothesis testing
- random sample
- match scores
- statistical power
- pac learning
- worst case
- small samples
- random samples
- likelihood ratio test
- sampling algorithm
- data sets
- statistical significance
- supervised learning
- machine learning