Appeal No. 1997-3225 Application No. 08/351,218 through 72 and 79 through 84. Claims 8, 10, 38 through 40, 53, 55 and 73 through 78 were canceled. The invention relates to a technique for allowing a computer to simulate an animated image of a human speaking. As identified in the specification on pages 9 and 10, a spoken language is broken down into units of speech, phonemes and diphthongs. Then as identified on page 10 of the specification, a video of a person speaking is obtained and individual frames which best represent the person using these units of speech are determined. As identified on pages 11 through 14 of the specification, these frames are then saved in a database, each image corresponding to a phoneme. As described on pages 14 and 15 of the specification, features in each image in the database are matched to identical features in the other images. These features in the images are called tiepoints. As identified on page 15 of the specification, an audio speech sequence to be synchronized with the images is then analyzed to determine spoken phonemes and their relative timing. Then as identified on pages 15 through 18 and 32 2Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 NextLast modified: November 3, 2007