recognizer.process(visionImage) { result, error in
  guard error == nil, let result = result else {
    // ...
    return
  }

  // Text recognition results
  let resultText = result.text
  // `VisionText` is made of `VisionTextBlock`s...
  for block in resultText.blocks {
    // ...and each `VisionTextBlock` has
    // text, confidence level, list of recognized languages and corner points and frame in which the text was detected...
    let blockText = block.text
    let blockConfidence = block.confidence
    let blockLanguages = block.recognizedLanguages
    let blockCornerPoints = block.cornerPoints
    let blockFrame = block.frame
    // ...and each `VisionTextBlock` is made of `VisionTextLine`s...
    for line in block.lines {
        // ...which again have text, confidence level, list of recognized languages, corner points and the frame in which the text was detected...
        let lineText = line.text
        let lineConfidence = line.confidence
        let lineLanguages = line.recognizedLanguages
        let lineCornerPoints = line.cornerPoints
        let lineFrame = line.frame
        // ... and each `VisionTextLine` is further made of `VisionTextElement`s...
        for element in line.elements {
            // ...and each `VisionTextElement` has
            // text, confidence level, list of recognized languages, corner points and the frame in which the text was detected.
            let elementText = element.text
            let elementConfidence = element.confidence
            let elementLanguages = element.recognizedLanguages
            let elementCornerPoints = element.cornerPoints
            let elementFrame = element.frame
        }
    }
  }
}