tony@wldrdg.UUCP (Tony Andrews) (11/12/86)
I've been disassembling the speech synthesizer that James Turner posted recently. My goal is to come up with a set of routines that can be linked into programs to support speech output. In the process of doing that, I've discovered a few features of the program that may be of interest: 1. The program supports two input modes. On startup, you're in english text mode. By typing a single period followed by return, you enter "phoneme mode". In this mode you can enter arbitrary phoneme strings using the same phoneme codes that you see echoed in english mode. Also, each phoneme may be followed by a pitch code (1-9). 2. Typing a backslash and return causes the last input line to be reused. 3. An input line of the form "%ddd" (where ddd is a decimal number) changes the rate at which speech is generated. The rate value can range from 20 through 127. 4. An input line of the form "!ddd" changes the base level pitch of the output. Again, numbers from 20 through 127 are accepted, but numbers less than around 80 seem to cause the output code to fail. There is a copyright notice in the binary that reads (sic): MC68000/AY-3-8910 SPEECH SYNTHERSIZER V:2.0 Copyright 1986 A.D.BEVERIDGE & M.N.DAY ALL RIGHTS RESERVED. If anyone knows these people or knows for sure that this isn't being sold as a product somewhere, please send me mail. I plan to post any speech output routines that I come up with, but I'd rather know the status of this program before I do that. Tony Andrews ...!ihnp4!onecom!wldrdg!tony Wildridge Consulting, Inc. Boulder, CO