{"id":3254,"date":"2024-08-15T14:48:37","date_gmt":"2024-08-15T11:48:37","guid":{"rendered":"https:\/\/docs.unibell.tech\/?page_id=3254"},"modified":"2024-08-15T14:48:37","modified_gmt":"2024-08-15T11:48:37","slug":"recognition","status":"publish","type":"page","link":"https:\/\/docs.unibell.tech\/?page_id=3254","title":{"rendered":"Recognition"},"content":{"rendered":"\n
\"\"<\/a><\/figure>\n\n\n\n

The \u2018Recognition\u2019 block is used for recognition of a voice response and recording it into a variable. You can use this block for recognition of a subscriber\u2019s response and for transferring it to a server via HTTP.<\/p>\n\n\n\n

In the block settings you can set the time of a response expectancy (2 seconds by default) and set a variable to assign a result of the recognition to.<\/p>\n\n\n\n

The exit from this block is carried out after the waiting time for the response specified in the settings, or at the moment when the system detects the end of the subscriber’s speech, if the switch “Silence detection” is set (detection occurs after about 2-3 seconds).<\/p>\n\n\n\n

\"\"<\/a><\/figure>\n\n\n\n

There is also a language selection option. Clicking on the corresponding item opens a drop-down list containing all available languages.<\/p>\n\n\n\n

\"\"<\/a><\/figure>\n\n\n\n

Below the field of a variable you can find the settings \u2018Additional processing\u2019.<\/p>\n\n\n\n

To activate it choose the option \u2018Required\u2019 and in the drop-down list choose a content type that you need to extract from a subscriber\u2019s speech to save it into a variable.<\/p>\n\n\n\n

On the output side of the \u2018Recognition\u2019 block you can find two circles:<\/p>\n\n\n\n

    \n
  1. \u2018Processed\u2019 in case the recognition was successful.<\/li>\n<\/ol>\n\n\n\n

    The system returns verification code qc=0 (that means recognized sufficiently).<\/p>\n\n\n\n

      \n
    1. \u2018Not processed\u2019 in case for some reason recognition and assignment to a variable was not successful.<\/li>\n<\/ol>\n\n\n\n
      \"\"<\/a><\/figure>\n\n\n\n

      The system can detect from a subscriber\u2019s speech the following data types:<\/p>\n\n\n\n