| 8 | |
| 9 | |
| 10 | {{{ |
| 11 | We call this project as Interslice - to be released under Festvox (Alan |
| 12 | would have more comments). |
| 13 | |
| 14 | The basic idea of interslice is to automatically build synthetic voices |
| 15 | from large speech databases typically available from public domain such |
| 16 | as librivox.org and loudlit.org. |
| 17 | |
| 18 | Interslice comes with a segmentation tool capable to handling infinitely |
| 19 | large corpora and chunking them into utterances and *.lab files. |
| 20 | |
| 21 | You may also want to refer to Kishore Prahallad, Arthur R Toth and Alan |
| 22 | W Black,, /"Automatic Building of Synthetic Voices from Large |
| 23 | Multi-Paragraph Speech Databases " |
| 24 | <http://speech.iiit.net/%7Espeech/publications/interslice_v3.pdf>/, in |
| 25 | Proceedings of Interspeech, Antwerp, Belgium 2007. |
| 26 | http://speech.iiit.net/%7Espeech/publications/interslice_v3.pdf |
| 27 | |
| 28 | }}} |
| 29 | |