[Kotlin] STT (Speech-to-Text)

구글  STT  (Speech-to-Text)

---

AndroidManifest.xml

<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.RECORD_AUDIO" />

 

 

 

 

 

 

activity_main.xml

<?xml version="1.0" encoding="utf-8"?>
<androidx.constraintlayout.widget.ConstraintLayout
    xmlns:android="http://schemas.android.com/apk/res/android"
    xmlns:app="http://schemas.android.com/apk/res-auto"
    xmlns:tools="http://schemas.android.com/tools"
    android:layout_width="match_parent"
    android:layout_height="match_parent"
    tools:context=".MainActivity">

    <TextView
        android:id="@+id/tvResult"
        android:layout_width="wrap_content"
        android:layout_height="wrap_content"
        android:text="This is TextView Result"
        android:textSize="26dp"
        app:layout_constraintTop_toTopOf="parent"
        app:layout_constraintLeft_toLeftOf="parent"
        app:layout_constraintRight_toRightOf="parent"
        app:layout_constraintBottom_toBottomOf="parent"/>

    <Button
        android:id="@+id/btnStart"
        android:layout_width="wrap_content"
        android:layout_height="wrap_content"
        android:textAllCaps="false"
        android:text="start speech"
        android:textSize="18dp"
        app:layout_constraintTop_toBottomOf="@+id/tvResult"
        app:layout_constraintLeft_toLeftOf="parent"
        app:layout_constraintRight_toRightOf="parent"
        app:layout_constraintBottom_toBottomOf="parent"
        app:layout_constraintVertical_bias="0.1"/>

</androidx.constraintlayout.widget.ConstraintLayout>

activity_main.xml

 

 

 

 

 

 

MainActivity.kt

package com.jwsoft.kotlinproject

import android.Manifest
import android.content.Intent
import android.content.pm.PackageManager
import android.os.Build
import androidx.appcompat.app.AppCompatActivity
import android.os.Bundle
import android.speech.RecognitionListener
import android.speech.RecognizerIntent
import android.speech.SpeechRecognizer
import android.widget.Toast
import androidx.core.app.ActivityCompat
import androidx.core.content.ContextCompat
import kotlinx.android.synthetic.main.activity_main.*

class MainActivity : AppCompatActivity() {

    private lateinit var speechRecognizer: SpeechRecognizer
    private lateinit var recognitionListener: RecognitionListener

    override fun onCreate(savedInstanceState: Bundle?) {
        super.onCreate(savedInstanceState)
        setContentView(R.layout.activity_main)

        requestPermission()

        var intent = Intent(RecognizerIntent.ACTION_RECOGNIZE_SPEECH)
        intent.putExtra(RecognizerIntent.EXTRA_CALLING_PACKAGE, packageName)
        intent.putExtra(RecognizerIntent.EXTRA_LANGUAGE, "ko-KR")

        setListener()

        btnStart.setOnClickListener {
            speechRecognizer = SpeechRecognizer.createSpeechRecognizer(this)
            speechRecognizer.setRecognitionListener(recognitionListener)
            speechRecognizer.startListening(intent)
        }

    }


    private fun requestPermission() {
        if (Build.VERSION.SDK_INT >= 23 && ContextCompat.checkSelfPermission(this,
                Manifest.permission.RECORD_AUDIO) != PackageManager.PERMISSION_GRANTED) {

            // 거부해도 계속 노출됨. ("다시 묻지 않음" 체크 시 노출 안됨.)
            // 허용은 한 번만 승인되면 그 다음부터 자동으로 허용됨.
            ActivityCompat.requestPermissions(this,
                arrayOf(Manifest.permission.RECORD_AUDIO), 0)
        }
    }


    private fun setListener() {
        recognitionListener = object: RecognitionListener {

            override fun onReadyForSpeech(params: Bundle?) {
                Toast.makeText(applicationContext, "음성인식을 시작합니다.", Toast.LENGTH_SHORT).show()
            }

            override fun onBeginningOfSpeech() {

            }

            override fun onRmsChanged(rmsdB: Float) {

            }

            override fun onBufferReceived(buffer: ByteArray?) {

            }

            override fun onEndOfSpeech() {

            }

            override fun onError(error: Int) {
                var message: String

                when (error) {
                    SpeechRecognizer.ERROR_AUDIO ->
                        message = "오디오 에러"
                    SpeechRecognizer.ERROR_CLIENT ->
                        message = "클라이언트 에러"
                    SpeechRecognizer.ERROR_INSUFFICIENT_PERMISSIONS ->
                        message = "퍼미션 없음"
                    SpeechRecognizer.ERROR_NETWORK ->
                        message = "네트워크 에러"
                    SpeechRecognizer.ERROR_NETWORK_TIMEOUT ->
                        message = "네트워크 타임아웃"
                    SpeechRecognizer.ERROR_NO_MATCH ->
                        message = "찾을 수 없음"
                    SpeechRecognizer.ERROR_RECOGNIZER_BUSY ->
                        message = "RECOGNIZER가 바쁨"
                    SpeechRecognizer.ERROR_SERVER ->
                        message = "서버가 이상함"
                    SpeechRecognizer.ERROR_SPEECH_TIMEOUT ->
                        message = "말하는 시간초과"
                    else ->
                        message = "알 수 없는 오류"
                }
                Toast.makeText(applicationContext, "에러 발생 $message", Toast.LENGTH_SHORT).show()
            }

            override fun onResults(results: Bundle?) {
                var matches: ArrayList<String> = results?.getStringArrayList(SpeechRecognizer.RESULTS_RECOGNITION) as ArrayList<String>

                for (i in 0 until matches.size) {
                    tvResult.text = matches[i]
                }
            }

            override fun onPartialResults(partialResults: Bundle?) {

            }

            override fun onEvent(eventType: Int, params: Bundle?) {

            }

        }
    }

}

실행 후 버튼 클릭 후 말하면 텍스트 뷰에 반영됨.

성능은 좋지 않음.