| US 7,565,284 B2 | ||
| Acoustic models with structured hidden dynamics with integration over many possible hidden trajectories | ||
| Li Deng, Redmond, Wash. (US); Alejandro Acero, Redmond, Wash. (US); Dong Yu, Kirkland, Wash. (US); and Xiang Li, White Plains, N.Y. (US) | ||
| Assigned to Microsoft Corporation, Redmond, Wash. (US) | ||
| Filed on Mar. 01, 2005, as Appl. No. 11/71,904. | ||
| Claims priority of provisional application 60/625222, filed on Nov. 05, 2004. | ||
| Prior Publication US 2006/0100862 A1, May 11, 2006 | ||
| This patent is subject to a terminal disclaimer. | ||
| Int. Cl. G10L 15/14 (2006.01) | ||
| U.S. Cl. 704—206 | 18 Claims |
| 1. A method of identifying a sequence of vocal tract resonance (VTR) distributions for a sequence of phonetic units, the method
comprising:
identifying a sequence of target distributions for a VTR; and
applying the sequence of target distributions to a finite impulse response filter to produce a sequence of VTR distributions,
utilizing the sequence of VTR distributions in a speech recognition process.
|