摘要 |
There are provided a data processing device and a data processing device control program which can appropriately generate an acoustic model for an unspecified speaker and an acoustic model for unspecified speaker most appropriate for a specified personal speech, considering distribution of feature parameters changing according to particular conditions such as speaker type, utterance vocabulary, utterance type, and utterance environment. A data processing device (1) includes a data classification section (1a), a data storage section (1b), a pattern model generation section (1c), a data control section (1d), a mathematical distance calculation section (1e), a pattern model conversion section (1f), a pattern model display section (1g), a region dividing section (1h), a divided content modification section (1i), a region selection section (1j), and a particular pattern model generation section (1k). |