I am working on a script that will have quite a bit of spoken text in it. I would like to keep the various "speech" files sotred as Text files, like speech1.txt. The file would contain the words for Agent to speak.
Question 1: Can I use the command hs.speak speech1.txt? If not, what is the best way to do this? Note: some of the text files will be somewhat long - >256 chars.
Question 2. I know I can embed MS Agent commands in the speech files, /Cmd xxx/, but already I have found that the AT&T voices operate quite a bit differently. What speech variables can I embed if I use the AT&T engine? I know Natural Voices can read files, but it gets back to the same question, how do I pipe the text file to it?
I have also been discovering another anomoly, using the period (.) or comma (,). In the MS Agent version, these are good for pauses. In Naturally Speaking, these get spoken, if there is more than one. Sometimes, they get spoken if there is only one. Has anybody figured out the rules for the AT&T version?
(I guess I had three questions. Oh well...)
Dennis [img]/infopop/emoticons/icon_cool.gif[/img]
Question 1: Can I use the command hs.speak speech1.txt? If not, what is the best way to do this? Note: some of the text files will be somewhat long - >256 chars.
Question 2. I know I can embed MS Agent commands in the speech files, /Cmd xxx/, but already I have found that the AT&T voices operate quite a bit differently. What speech variables can I embed if I use the AT&T engine? I know Natural Voices can read files, but it gets back to the same question, how do I pipe the text file to it?
I have also been discovering another anomoly, using the period (.) or comma (,). In the MS Agent version, these are good for pauses. In Naturally Speaking, these get spoken, if there is more than one. Sometimes, they get spoken if there is only one. Has anybody figured out the rules for the AT&T version?
(I guess I had three questions. Oh well...)
Dennis [img]/infopop/emoticons/icon_cool.gif[/img]
Comment