[orca: 1/2] speech: respect SSML norm

[Joanmarie Diggs <joanied src gnome org>]

commit 3f253a3258586c71d422584b9d00e451666aac56
Author: Samuel Thibault <samuel thibault ens-lyon org>
Date:   Fri Jul 6 18:07:28 2018 +0200

    speech: respect SSML norm
    
    Some speech syntheses require to have SSML input enclosed in a speak tag, as
    SSML says.

 src/orca/speechdispatcherfactory.py | 2 ++
 1 file changed, 2 insertions(+)
---
diff --git a/src/orca/speechdispatcherfactory.py b/src/orca/speechdispatcherfactory.py
index e7d68b629..793886101 100644
--- a/src/orca/speechdispatcherfactory.py
+++ b/src/orca/speechdispatcherfactory.py
@@ -369,6 +369,8 @@ class SpeechServer(speechserver.SpeechServer):
         #
         text = text.replace('\n.', '\n')
 
+        text = "<speak>" + text + "</speak>"
+
         self._apply_acss(acss)
         self._debug_sd_values("Speaking '%s' " % text)
         self._send_command(self._client.speak, text, **kwargs)