AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

To Voice or Not To Voice
 
 

I think I’m barking up a dead tree, but I’ll try anyway.

Occasionally, when I try adding an ability to my Pandorabot, it interferes with some feature that I want to keep.

For example, the single most requested ability people request of my bot is that it reply with a voice.  However, my bot will show images, or play music files and videos.  In the past, when I’ve experimented with adding text-to-speech, the bot seems foolish as it reads aloud the URLs and the accompanying code.

Is there a way to isolate portions of a response that I don’t want produced as sound as though it were text?

 

 
  [ # 1 ]

I haven’t had anyone complain about reading urls. It is really the only way to do it.
But, if you are sending a raw text stream, you need to strip commands that may be embedded before you pass it to the tts engine.
You could use a regular expression to strip the unwanted items.
The alternative is to pass just what you want to the tts engine via a function.
Say(“hi, how are you”)

 

 
  login or register to react