macaroni

analyze

Description:
The analyze executor is used to analyze an image (from clipboard or file) using the Azure Computer Vision service.

Parent: execute, then, else

Value (simple form): string specifying the image to be analyzed
Value (key/value form):

• analyze: string, required, specifying the input file name, or {clipboard.image} of the image to analyze
• feature: string, optional, specifying the visual feature(s) to analyze in the image (default: TextRegions)
• language: string, optional, specifying the language used for descriptions and hint for OCR (default: from configuration)
• endpoint: string, optional, specifying the Azure Computer Vision endpoint (default: from configuration)
• key: string, optional, specifying the Azure Computer Vision api-key (default: from configuration)

# simple form
- analyze: (string)

# simple example
- analyze: {clipboard.image}

# key/value form
- analyze: (string, required)
  feature: (string, optional)
  language: (string, optional)
  endpoint: (string, optional)
  key: (string, optional)

# key/value example
- analyze: '{folders.desktop}\image.png'
  feature: 'Text'

EXAMPLE

# describe image on clipboard using image analysis service
commands:
- name: describe image on clipboard using image analysis service
  triggers:
  - phrase: describe that
  - phrase: tell me about that
  execute:
  - analyze: '{clipboard.image}'
    feature: Description
  - display: '{analyze.text}'
  - clipboard: '{analyze.text}'
  - speak: '{analyze.text}'

Additional information:

• After the user speaks a phrase, subsequent executors in the same execution context can access the analyzed result:
  • analyze.text - the text recognized from the image
  • analyze.result - the Vision SDK ImageAnalysisResult json object (see json executor parse this result)

Status: ✅ IMPLEMENTED