Hi Tom, Full agreement here. We have started using SAPI in our new game engine as well, as has been evident in ChangeReaction 2 and SilverDollar. We use a very simple formula for deciding what should or should not be presented with TTS. That is, if this was a video game and the information being presented would be text on the screen, that information should be presented with TTS. Otherwise, human voices should be used.
In the main menus for both CR and SD, we do have a human voice for the main options, because we are thinking of these options more as icons. Likewise, in CR, the coins in a video game would be graphically represented, hence they are named by a human voice during game play. We resisted using TTS for a long time. In fact, we weren't using it at all in early betas of CR2. In the end, the dubious benefits are far outweighed by the drawbacks of using human voices for every bit of textual output. On May 4, 2013, at 11:07 AM, Thomas Ward <thomasward1...@gmail.com> wrote: > Hi everyone, > > This morning I had some time on my hands so I began really working on > Mysteries of the Ancients, the new version, and decided to add SAPI 5 > support to help expedite the development process. What I discovered > when I compiled and ran this very early prototype is that the SAPI > support works far better than any release before, and I realized to my > amazement it really is the best solution for speaking menus, status > messages, etc for a number of reasons. > > 1. It takes less development time to call a single function to pass a > string of text to a SAPI speak function than it does to record, edit, > load, and play a speech clip. Basically, if you got a game idea in > mind you can really speed up the process just by using the SAPI voices > installed on the machine. > > 2. It saves drive space, takes up less CPU power, and less memory. In > the last beta of Mysteries of the Ancients,beta 22, the Speech > directory was 36 MB. Now, of course, using SAPI it will take up less > drive space as well as be a smaller download. I also noticed looking > at the over all system performance that it uses less memory so that's > another reason to use SAPI. > > 3. Its customizable. This probably goes without saying, but using SAPI > voices you can change the voice, pitch, rate, and volume of the > default voice in the Control Panel meaning if you don't like the > default voice you can always switch to a different one or buy a better > one. > > This was always a bit of a problem in prior test releases. The voice > was too fast, too slow, it had an accent, or whatever and I was going > crazy trying to find a voice that everyone liked. So using SAPI you, > the end user, can set it up anyway you like in the Windows Control > Panel. > > 4. Its very inexpensive. If you hire voice actors to read the menus, > status messages, items, etc that can cost lots of money not to mention > require time to edit and use. Even if a developer uses Acapela Heather > or Nuance Tom legally the developer is suppose to pay royalties for > the use of that voice in a commercial product. This way the developer > isn't using the voices directly and the customer fits the bill for any > high quality voices like Ivona, Nuance TTS, Cepstral, etc. > > 5. Its very dynamic and flexible. With prerecorded speech there is > always the problem of speaking dynamic content like a player's name or > speaking a message with a number of variables involved. Here you just > create a message string using the dynamic content, give it to sAPI, > and forget about it. > > For example, in my wrestling game there are a number of messages of > play by play action like this. "John Cena quickly moves forward, picks > up the Rock, and slams him to the mat." In a case like this the speech > output has to be flexible enough to use the same message from match to > match only substituting the name of the performers in the match. You > can't really do that easily with prerecorded speech clips, but you can > easily do it with SAPI just by using a couple of string variables in > place of the performers name. Therefore it doesn't matter who is in > the match it will always speak the information no matter what, and the > developer doesn't have to spend a month writing a bunch of if > statements to load this or that speech clip on demand. :D > > So I think after seeing the results in person I think I am going to be > using SAPI output from now on in my Windows games. I only regret I > didn't do this earlier. I could have saved myself a lot of wasted time > and energy trying to make prerecorded speech clips. > > Cheers! > > --- > Gamers mailing list __ Gamers@audyssey.org > If you want to leave the list, send E-mail to gamers-unsubscr...@audyssey.org. > You can make changes or update your subscription via the web, at > http://audyssey.org/mailman/listinfo/gamers_audyssey.org. > All messages are archived and can be searched and read at > http://www.mail-archive.com/gamers@audyssey.org. > If you have any questions or concerns regarding the management of the list, > please send E-mail to gamers-ow...@audyssey.org. --- Gamers mailing list __ Gamers@audyssey.org If you want to leave the list, send E-mail to gamers-unsubscr...@audyssey.org. You can make changes or update your subscription via the web, at http://audyssey.org/mailman/listinfo/gamers_audyssey.org. All messages are archived and can be searched and read at http://www.mail-archive.com/gamers@audyssey.org. If you have any questions or concerns regarding the management of the list, please send E-mail to gamers-ow...@audyssey.org.
