I’m a Korean user experiencing an issue with the Play Prompt function. I set a variable with my Korean name ‘동진욱’ (Dong Jin-uk), but there seems to be an issue with variable playback. Instead of playing my name that was assigned to the variable, it’s literally speaking the variable name ‘$name’. As a Korean language user, how can I make the Play Prompt function correctly read my actual Korean name value instead of the variable syntax
I noticed something different between Korean and English language settings in the Play Prompt editor. When I select English as the language, I can see a ‘Text’ option in the dropdown menu where I can input variable names. However, this ‘Text’ option doesn’t appear when Korean is selected as the language. Does this mean that variable playback is not supported for Korean language prompts? Or is there a different way to implement variables in Korean prompts?
Ultimately, we want to play back the values selected from the database in Korean.
Hi Dong Jin-uk,
When you type text for “Voice” segment types, a prompt-generation TTS is used to pre-create prompt files during IVR scenario development, so you won’t need to record the voice prompts yourself. This is why variables are not expanded there - the conversion is done once for each scenario, not for each invocation instance.
“Text” segment type, though, is dynamic, and it does take variables as input.
If you have TTS integration (looks like you do) and it does not show “Text” selection for a segment type for Korean, ask support to enable it on your tenant/cluster (there is a configuration table for a cluster that defines which languages are enabled for TTS).
I hope this helps,
Sergey
type “voice” is used for dual purpose - it plays the prerecorded file (if any) or annotation text if the files is not uploaded. So the content of the file can’t have variables - the same as for the real prerecorded file. And in order to add valiables into this prompt you must use “add segment”
In other words the prompt like: My Name is $(name). - should be split into 2 segments:
- My Name (type voice)
- $(name). (type Voice from Parameter)
The whole concept of Voice type was around prerecorded files mostly
Today the TTS quality has got to the human level quality, customers are switching to Text type.
Text is not showing up in your case because it’s not enabled in your system
What system do you use - is it your private installation or BP cloud ? If that is a BP cloud we can turn it on Quickly, if not i’ll have to give you a query to run to change the config setting for the cluster
Thank you Sergey for the clear explanation
After changing the value of the TTS column to 1 in the languages table of sp_config DB, the text was displayed properly and variable playback became possible. Thank you.
