Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Skill Type

Status
colourGreen
titlePOST PROCESS

Overview

Excerpt

The

Pronunciation Skill enables the DP to change their

pronunciation

of words based on a simple string match/replacement

The Pronunciation Skill can be included with any DDNA Studio project as a post-process skill. This skill enables the user to provide skill allows you to set up a list of words or phrases that require updating without making updates to the base NLP or Gen AI conversation.

Sample Skill Definition

Code Block
{
  "name": "Text Replacement Skill",
  "summary": "Create a list of words to update in post process",
  "description": "Create a list of words to update in post process",
  "isPublic": false,
  "status": "ACTIVE",
  "serviceProvider": "SKILL_API",
  "category": null,
  "endpointInitialize": null,
  "endpointSession": null,
  "endpointExecute": "https://us-central1-text-replacement-skill.cloudfunctions.net/smTextReplace",
  "endpointEndSession": null,
  "endpointEndProject": null,
  "endpointMatchIntent": null,
  "languages": null,
  "config": {
    "skillType": "POST_PROCESS",
    "configMeta": [
      {
        "name": "replacementList",
        "type": "TEXT",
        "label": "List of text replacement updates",
        "required": true
      }
    ]
  }
}

The skill definition creates an replacementList input that allows you to paste a string of phrases and their replacements in the conversation. Phrase/Replacement specific pronunciations.

This skill is:

  • Built and owned by Soul Machines Ltd.

  • Only supported in English.

Limitations

  • Pronunciation replacement does not work when using the “Ask Digital Person to repeat...” text field in the preview window. Replacement is only done during an conversation.

  • Enclosing @pronounce in parentheses like (@pronounce()) doesn't get read correctly. The work workaround is to add a space like ( @pronounce() ).

Configurations

Provide the following information in the Pronunciation Skill configuration screen in Digital DNA Studio. See Adding Skills to your Digital Person for detailed instructions:

image-20240609-224907.pngImage Added

Field

Type

Description

REPLACEMENT LIST

Status
colourRed
titlerequired

TEXT

String of key word or phrase and their replacement in the conversation, separated by a vertical bar |. Multiple phrase/replacement pairs can be entered into the text area field

and are

delimited by a

new line (\n), while the individual words to replace are delimited by a dash (-)

comma.

ex: Sentence - string\nStoll - @pronounce(Stoll, Stall)

image-20240529-231709.pngImage Removed

Sample Request

Code Block
{
  "projectId": "ABC123",
  "deploymentEnvironment": "preview",
  "sessionId": "7600aff2-6ebd-44de-bcd2-1facd8ab4f29",
  "intent": {
    "name": "MY_INTENT",
    "confidence": 0
  },
  "text": "This is a sentence to demonstrate the text replacement skill, developed by David Stoll at Soul Machines.",
  "config": {
    "pronounceList": "Sentence - string\nStoll - @pronounce(Stoll, Stall)"
  }
}

Sample Response

Code Block{ "output": { "text": "This is a string to demonstrate the text replacement skill, developed by David @pronounce(Stoll, Stall) at Soul Machines.", "variables": {} }, "endConversation": false, "endRouting": false }

Example:

cm|centimeter, 888|triple eight

Contents

Table of Contents