If you want, AGS can pretend to 'read' out speech text. Go to the 'Lip Sync' pane in the editor.

Normally when a character is talking, AGS simply cycles through the speech animation from start to finish, and then loops back round. However, lip syncing allows you to be cleverer by specifying a particular frame to go with various letters and sounds. Then, as the character talks, AGS plays appropriate frames to simulate the character actually saying those words.

In the Lip Sync Editor, you have 20 text boxes, one for each possible frame of the talking loop. In each box, you can enter all the letters which will cause that frame of the loop to be played. Letter combinations such as 'th' and 'ch' can be used too - AGS will match each part of the spoken text to the longest possible phrase in the lip sync editor.

separate the letters by forward slashes. For example,

3  R/S/Th/G

will mean that frame 3 of the character's talking animation is shown whenever the letter R, S, Th or G is spoken.

The "Default frame for unlisted characters" box allows you to set which frame is used when a character not listed in any of the text boxes is encountered.

Voice speech lip sync

AGS supports lip syncing voice speech to the talking animation. If you enable this feature, you cannot use the standard lip-sync for non-voice lines.

NOTE: This is an unofficial feature and is not currently supported. Use at your own risk

NOTE: The voice sync feature only supports Sierra-style speech.

In order to do this, you need to download the third-party PAMELA application:

http://www-personal.monash.edu.au/~myless/catnap/pamela/

Set up the phenomes in Pamela so that there are only 10 (or as many talking frames as you have) available choices. Then, in the Lip Sync pane of AGS, change the Type property to "Voice". Enter the Pamela phenomes into the text boxes to create the association between the pamela phenome code and the AGS frame number.

For example, enter "AY0" into frame 0's box, "E" into frame 1, and so forth - corresponding to how it is set up in Pamela. For multiple phenomes to share the same frame, seperate them with forward slashes -- for example, "AY0/AY1" allows both of those phenomes to correspond to the specified frame.

Use the Pamela application on each of your speech lines, and save a Pamela project file (.pam file) for each speech file, naming it the same as the speech.

For example, the pamela project for EGO46.OGG would be called EGO46.PAM, placed in your game's Speech folder.

When you build the game, this pam file is compiled into the speech.vox and will be used to sync the animation of the talking frames during the game.

NOTE: Voice lip sync does not work well with MP3 files. It is strongly recommended that you use OGG or WAV for speech.