Scroll export button | ||||||||
---|---|---|---|---|---|---|---|---|
|
Info |
---|
This guide explains in detail each Dialplan application and provides practical examples and use cases. Updated: July August 2023 Permalink: https://wildix.atlassian.net/wiki/x/ahDOAQ |
Table of Contents |
---|
Basic information on how Dialplan works: Dialplan: How does it work?
...
Note |
---|
Notes:
|
Dialplan application Speech to text allows to create an automatic system that ask the caller some questions, recognizes the answers, and lets you save the recorded files.
...
- RECOGNITION_STATUS = SUCCESS / NOT_RECOGNIZED / EMPTY_FILE / SERVICE_FAIL
- RECOGNITION_RESULT = returns result of the speech recognition, e.g. ”My name is Alex”
- RECOGNITION_AUDIO = returns name of the audio file, e.g.:
/dev/shm/var/spool/callweaver/monitor/stt/20220916-170648-+380487770777-wildixbox-1663340800.21-1-name.wav
Where:
a) “20220916-170648” = date and time
b) “+380487770777” = Caller ID
c) “wildixbox-1663340800.21” = Linked ID, which is a unique variable of the call (can be found in CDR)
d) “1” = number of question in the Dialplan (1- for the first question, 2- for the second and so on)
e) “name” = variable of the RECOGNITION_LABEL, if set RECOGNITION_RESULTS = a separate channel variable that stores JSON with all the recognition results for the current channel
Note Note: The maximum length of the variable RECOGNITION_RESULTS is 4092 Bytes. The data that exceeds this limit, cannot be saved to RECOGNITION_RESULTS.
Example:
Code Block |
---|
RECOGNITION_RESULTS = { "questions":[ { "question":"Hello, what is your name?", "status":"SUCCESS", "audio_file":"/rw2/backups/20220916-170648-+380487770777-wildixbox-1663340800.21-1-name.wav", "result":"i'm alex", "label":"name" }, { "question":"What is your age?", "status":"NOT_RECOGNIZED", "audio_file":"/rw2/backups/20220916-170706-+380487770777-wildixbox-1663340800.21-2-age.wav", "result":"", "label":"age" } ], "timestamp":"2022-09-16 17:06:48", "linkedid":"wildixbox-1663340800.21" } |
...