Select the data type and dialect, then specify the metrics and variables required, along with the range of values. You can also specify whether you want the behavioural data and sound files.
For these talkers:
For these dialects: