Analyzing speech and doing an action - swift

Question

I am trying to analyze a sentence and make an action depending on that.
Here are examples of sentences received from speech recognition. I wrote couple of sentences, because we actually don't know what user is going to say for sure, and if he at all says the right pattern.
var str = "20 minutes to take a shower"
var sentence = "seven minutes to make 10 last homeworks"
var sentence2 = "strum guitar for 15 minutes"
var plans = "launch with friend 12:15, then drawing lesson"
I want to extract "20 minutes" ; assign it for the timeValue and launch the timer.
Also, to assign a task to taskValue to represent the task that I am doing. (I thought getting task value by removing "20 minutes" from the initial sentence).
What do you think is the best way to work with the String that I need to analyze?
I thought of finding index ranges and then cutting/copying with the help of indexes, but
The format of the indexes it returns is like this: and I don't know how to extract the number of index. (In this case: 0 through 10)
<NSSimpleRegularExpressionCheckingResult: 0x6000027f0140>{0, 10}{<NSRegularExpression: 0x600003cfcb40> [0-9]{1,} minutes 0x1}
How to ignite the timer? We have to validate that there's the proper command given. And, when we get the timeValue and taskValue back, then how do we ignite the timer? (The whole process was: User pushes button -> user speaking -> speech recognized(and displayed on the screen label) -> sentence analyzed(?) -> timer starts(?) and task displayed in the label(?) )
What is your recommendation for architecture of speech analysis system. Maybe you know some articles on this topic?
Here's the logic for the speech detection.
var timeValue: Int = Int()
var taskValue: String = ""
func stringDeduction(of inputText: String) -> (Int?, String?) {
let pattern = "[0-9]{1,} minutes"
let regexOptions: NSRegularExpression.Options = [.caseInsensitive]
let matchingOptions: NSRegularExpression.MatchingOptions = [.reportCompletion]
// TODO - catch errors with regex
let regex = try! NSRegularExpression(pattern: pattern, options: regexOptions)
// } catch {
// print("error in regex")
// }
let range = NSRange(location: 0, length: inputText.utf8.count)
// \d - matches any digit
// Pattern for time format like this 00:00
//let patternForTime = "[0-9]{1,}:[0-9]{1,2}"
if let matchIndex = regex.firstMatch(in: inputText, options: matchingOptions, range: range) {
print(matchIndex)
} else {
print("No match.")
}
// check whether the string matches and print one of two messages
if let match = regex.firstMatch(in: inputText, range: NSRange(location: 0, length: inputText.utf8.count)) {
print("*: Match!")
} else {
print("*: No match.")
}
/* Question - how to use "mathces" properly?!
if let match = regex.matches(in: testString, options: .reportCompletion ,range: NSRange(location: 0, length: testString.utf8.count)) {
print("*: Match!")
print(match)
} else {
print("*: No match.")
}
*/
func matches(for regex: String, in text: String) -> [String] {
do {
let regex = try NSRegularExpression(pattern: regex)
let results = regex.matches(in: text, options: [], range: NSRange(text.startIndex..., in: text))
return results.map {
String(text[Range($0.range, in: text)!])
}
} catch {
print("invalid regex")
return []
}
}
var task = regex.stringByReplacingMatches(in: inputText, options: .withoutAnchoringBounds, range: range, withTemplate: "")
//var taskMutableString = NSMutableString(string: str)
//regex.replaceMatches(in: taskMutableString, options: .withoutAnchoringBounds, range: range, withTemplate: "")
//taskMutableString
var timeStringArray = matches(for: pattern, in: inputText)
var timeString = timeStringArray[0]
let time = Int(timeString.replacingOccurrences(of: " minutes", with: ""))
timeValue = time ?? Int()
taskValue = task.replacingOccurrences(of: "to" , with: "", options: .caseInsensitive, range: task.startIndex..<task.index(task.startIndex, offsetBy: 4))
let taskReturn = taskValue
return (time, taskReturn)
// the regex ^[ \t]+|[ \t]+$ matches excess whitespace at the beginning or end of a line.
// what regex, or string method matches exess whitespace at the beginning of a line
}
Here's the extension to work with string like with array. Like this, - str[0..2]
extension String {
subscript (bounds: CountableClosedRange<Int>) -> String {
let start = index(startIndex, offsetBy: bounds.lowerBound)
let end = index(startIndex, offsetBy: bounds.upperBound)
return String(self[start...end])
}
subscript (bounds: CountableRange<Int>) -> String {
let start = index(startIndex, offsetBy: bounds.lowerBound)
let end = index(startIndex, offsetBy: bounds.upperBound)
return String(self[start..<end])
}
}
Here's another approach I have tried: (Although it doesn't seem good)
let timeArray = ["one minute", "two minutes", "three minutes", "four minutes", "five minutes", "six minutes", "seven minutes", "eight minutes", "nine minutes"]
let timeValueCheck = "Answer: \(sentence.containsAny(of: timeArray) ?? "doesn't contain")"
//Dividing String into words
let abc: [String] = str.components(separatedBy: " ")
// Finding a number within an array of words
// There's a problem if there are couple of numbers in the sentence, it returns all of them and not only the needed time.
let numbers = abc.compactMap {
// convert each substring into an Int
return Int($0)
}
for i in 1...100 {
if str.contains(String(i) + " minutes") {
print(i)
}
}
Thank you for any of your help! I've just gone insane with this task for 3 months. Also, if something is unclear or a bit messy, please tell me! I'll try to correct.