How do I convert voice to text in iOS [closed] - ios

Question

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 5 years ago.
Improve this question
As far as I know , apple native framework doesn't have APIs for converting voice to text and we have to go for third party framework to do that and it has so many drawbacks like user has to microphone to convert from voice to text.
But I can find lots of information for converting text to voice but not the other way
Couldn't find any clear information about this and mostly it has so many uncertain things.
If someone could shed some light it'd be really great !

Answer 1

For Objective C, I wrote a Speech Converter class a while back to convert voice to text.
Step 1: Create A Speech Convertor class
Create a new Cocoa Class and subclass it from NSObject.
Name it let's say ATSpeechRecognizer.
In ATSpeechRecognizer.h:
#import <Foundation/Foundation.h>
#import <Speech/Speech.h>
#import <AVFoundation/AVFoundation.h>
typedef NS_ENUM(NSInteger, ATSpeechRecognizerState) {
ATSpeechRecognizerStateRunning,
ATSpeechRecognizerStateStopped
};
#protocol ATSpeechDelegate<NSObject>
#required
/*This method relays parsed text from Speech to the delegate responder class*/
-(void)convertedSpeechToText:(NSString *) parsedText;
/*This method relays change in Speech recognition ability to delegate responder class*/
-(void) speechRecAvailabilityChanged:(BOOL) status;
/*This method relays error messages to delegate responder class*/
-(void) sendErrorInfoToViewController:(NSString *) errorMessage;
#optional
/*This method relays info regarding whether speech rec is running or stopped to delegate responder class. State with be either ATSpeechRecognizerStateRunning or ATSpeechRecognizerStateStopped. You may or may not implement this method*/
-(void) changeStateIndicator:(ATSpeechRecognizerState) state;
#end
#interface ATSpeechRecognizer : NSObject <SFSpeechRecognizerDelegate>
+ (ATSpeechRecognizer *)sharedObject;
/*Delegate to communicate with requesting VCs*/
#property (weak, nonatomic) id<ATSpeechDelegate> delegate;
/*Class Methods*/
-(void) toggleRecording;
-(void) activateSpeechRecognizerWithLocaleIdentifier:(NSString *) localeIdentifier andBlock:(void (^)(BOOL isAuthorized))successBlock;
#end
And in ATSpeechRecognizer.m:
#import "ATSpeechRecognizer.h"
#interface ATSpeechRecognizer ()
/*This object handles the speech recognition requests. It provides an audio input to the speech recognizer.*/
#property SFSpeechAudioBufferRecognitionRequest *speechAudioRecRequest;
/*The recognition task where it gives you the result of the recognition request. Having this object is handy as you can cancel or stop the task. */
#property SFSpeechRecognitionTask *speechRecogTask;
/*This is your Speech recognizer*/
#property SFSpeechRecognizer *speechRecognizer;
/*This is your audio engine. It is responsible for providing your audio input.*/
#property AVAudioEngine *audioEngine;
#end
#implementation ATSpeechRecognizer
#pragma mark - Constants
//Error Messages
#define kErrorMessageAuthorize #"You declined the permission to perform speech Permission. Please authorize the operation in your device settings."
#define kErrorMessageRestricted #"Speech recognition isn't available on this OS version. Please upgrade to iOS 10 or later."
#define kErrorMessageNotDetermined #"Speech recognition isn't authorized yet"
#define kErrorMessageAudioInputNotFound #"This device has no audio input node"
#define kErrorMessageRequestFailed #"Unable to create an SFSpeechAudioBufferRecognitionRequest object"
#define kErrorMessageAudioRecordingFailed #"Unable to start Audio recording due to failure in Recording Engine"
#pragma mark - Singleton methods
+ (ATSpeechRecognizer *)sharedObject {
static ATSpeechRecognizer *sharedClass = nil;
static dispatch_once_t onceToken;
dispatch_once(&onceToken, ^{
sharedClass = [[self alloc] init];
});
return sharedClass;
}
- (id)init {
if (self = [super init]) {
}
return self;
}
#pragma mark - Recognition methods
-(void) activateSpeechRecognizerWithLocaleIdentifier:(NSString *) localeIdentifier andBlock:(void (^)(BOOL isAuthorized))successBlock{
//enter Described language here
if([localeIdentifier length]>0){
NSLocale *locale = [[NSLocale alloc] initWithLocaleIdentifier:localeIdentifier];
_speechRecognizer = [[SFSpeechRecognizer alloc] initWithLocale:locale];
_speechRecognizer.delegate = self;
_audioEngine = [[AVAudioEngine alloc] init];
[self getSpeechRecognizerAuthenticationStatusWithSuccessBlock:^(BOOL isAuthorized) {
successBlock(isAuthorized);
}];
}
else{
successBlock(NO);
}
}
/*Microphone usage Must be authorized in the info.plist*/
-(void) toggleRecording{
if(_audioEngine.isRunning){
[self stopAudioEngine];
}
else{
[self startAudioEngine];
}
}
#pragma mark - Internal Methods
/*
In case different buttons are used for recording and stopping, these methods should be called indiviually. Otherwise use -(void) toggleRecording.
*/
-(void) startAudioEngine{
if([self isDelegateValidForSelector:NSStringFromSelector(#selector(changeStateIndicator:))]){
[_delegate changeStateIndicator:ATSpeechRecognizerStateRunning];
}
[self startRecordingSpeech];
}
-(void) stopAudioEngine{
if([self isDelegateValidForSelector:NSStringFromSelector(#selector(changeStateIndicator:))]){
[_delegate changeStateIndicator:ATSpeechRecognizerStateStopped];
}
[_audioEngine stop];
[_speechAudioRecRequest endAudio];
self.speechRecogTask = nil;
self.speechAudioRecRequest = nil;
}
/*
All the voice data is transmitted to Apple’s backend for processing. Therefore, it is mandatory to get the user’s authorization. Speech Recognition Must be authorized in the info.plist
*/
-(void) getSpeechRecognizerAuthenticationStatusWithSuccessBlock:(void (^)(BOOL isAuthorized))successBlock{
[SFSpeechRecognizer requestAuthorization:^(SFSpeechRecognizerAuthorizationStatus status) {
switch (status) {
case SFSpeechRecognizerAuthorizationStatusAuthorized:
successBlock(YES);
break;
case SFSpeechRecognizerAuthorizationStatusDenied:
[self sendErrorMessageToDelegate:kErrorMessageAuthorize];
successBlock(NO);
case SFSpeechRecognizerAuthorizationStatusRestricted:
[self sendErrorMessageToDelegate:kErrorMessageRestricted];
successBlock(NO);
case SFSpeechRecognizerAuthorizationStatusNotDetermined:
[self sendErrorMessageToDelegate:kErrorMessageNotDetermined];
successBlock(NO);
break;
default:
break;
}
}];
}
-(void) startRecordingSpeech{
/*
Check if the Task is running. If yes, Cancel it and start anew
*/
if(_speechRecogTask!=nil){
[_speechRecogTask cancel];
_speechRecogTask = nil;
}
/*
Prepare for the audio recording. Here we set the category of the session as recording, the mode as measurement, and activate it
*/
AVAudioSession *audioSession = [AVAudioSession sharedInstance];
#try {
[audioSession setCategory:AVAudioSessionCategoryRecord error:nil];
[audioSession setMode:AVAudioSessionModeMeasurement error:nil];
[audioSession setActive:YES error:nil];
} #catch (NSException *exception) {
[self sendErrorMessageToDelegate:exception.reason];
}
/*
Instantiate the recognitionRequest. Here we create the SFSpeechAudioBufferRecognitionRequest object. Later, we use it to pass our audio data to Apple’s servers.
*/
#try {
_speechAudioRecRequest = [[SFSpeechAudioBufferRecognitionRequest alloc] init];
} #catch (NSException *exception) {
[self sendErrorMessageToDelegate:kErrorMessageRequestFailed];
}
/*
Check if the audioEngine (your device) has an audio input for recording.
*/
if(_audioEngine.inputNode!=nil){
AVAudioInputNode *inputNode = _audioEngine.inputNode;
/*If true, partial (non-final) results for each utterance will be reported.
Default is true*/
_speechAudioRecRequest.shouldReportPartialResults = YES;
/*Start the recognition by calling the recognitionTask method of our speechRecognizer. This function has a completion handler. This completion handler will be called every time the recognition engine has received input, has refined its current recognition, or has been canceled or stopped, and will return a final transcript.*/
_speechRecogTask = [_speechRecognizer recognitionTaskWithRequest:_speechAudioRecRequest resultHandler:^(SFSpeechRecognitionResult * _Nullable result, NSError * _Nullable error) {
BOOL isFinal = NO;
if(result!=nil){
if([self isDelegateValidForSelector:NSStringFromSelector(#selector(convertedSpeechToText:))]){
[_delegate convertedSpeechToText:[[result bestTranscription] formattedString]];
}
isFinal = [result isFinal]; //True if the hypotheses will not change; speech processing is complete.
}
//If Error of Completed, end it.
if(error!=nil || isFinal){
[_audioEngine stop];
[inputNode removeTapOnBus:0];
self.speechRecogTask = nil;
self.speechAudioRecRequest = nil;
if(error!=nil){
[self stopAudioEngine];
[self sendErrorMessageToDelegate:[NSString stringWithFormat:#"%li - %#",error.code, error.localizedDescription]];
}
}
}];
/* Add an audio input to the recognitionRequest. Note that it is ok to add the audio input after starting the recognitionTask. The Speech Framework will start recognizing as soon as an audio input has been added.*/
AVAudioFormat *recordingFormat = [inputNode outputFormatForBus:0];
[inputNode installTapOnBus:0 bufferSize:1024 format:recordingFormat block:^(AVAudioPCMBuffer * _Nonnull buffer, AVAudioTime * _Nonnull when) {
[self.speechAudioRecRequest appendAudioPCMBuffer:buffer];
}];
/*Prepare and start the audioEngine.*/
[_audioEngine prepare];
#try {
[_audioEngine startAndReturnError:nil];
} #catch (NSException *exception) {
[self sendErrorMessageToDelegate:kErrorMessageAudioRecordingFailed];
}
}
else{
[self sendErrorMessageToDelegate:kErrorMessageAudioInputNotFound];
}
}
-(BOOL) isDelegateValidForSelector:(NSString*)selectorName{
if(_delegate!=nil && [_delegate respondsToSelector:NSSelectorFromString(selectorName)]){
return YES;
}
return NO;
}
-(void) sendErrorMessageToDelegate:(NSString*) errorMessage{
if([self isDelegateValidForSelector:NSStringFromSelector(#selector(sendErrorInfoToViewController:))]){
[_delegate sendErrorInfoToViewController:errorMessage];
}
}
#pragma mark - Speech Recognizer Delegate Methods
-(void) speechRecognizer:(SFSpeechRecognizer *)speechRecognizer availabilityDidChange:(BOOL)available{
if(!available){
[self stopAudioEngine];
}
[_delegate speechRecAvailabilityChanged:available];
}
And that's it. Now you can use this class anywhere in any project you want to convert voice to text. Just be sure to read the guidance comments if you feel confused about how it works.
Step 2: Set up the ATSpeechRecognizer Class in your VC
Import ATSpeechRecognizer in your View Controller and set up the delegate like this:
#import "ATSpeechRecognizer.h"
#interface ViewController : UIViewController <ATSpeechDelegate>{
BOOL isRecAllowed;
}
Use the following method on VC viewDidLoad to set it up and running:
-(void) setUpSpeechRecognizerService{
[ATSpeechRecognizer sharedObject].delegate = self;
[[ATSpeechRecognizer sharedObject] activateSpeechRecognizerWithLocaleIdentifier:#"en-US" andBlock:^(BOOL isAuthorized) {
isRecAllowed = isAuthorized; /*Is operation allowed or not?*/
}];
}
Now set up delegate methods:
#pragma mark - Speech Recog Delegates
-(void) convertedSpeechToText:(NSString *)parsedText{
if(parsedText!=nil){
_txtView.text = parsedText; //You got Text from voice. Use it as you want
}
}
-(void) speechRecAvailabilityChanged:(BOOL)status{
isRecAllowed = status; //Status of Conversion ability has changed. Use Status flag to allow/stop operations
}
-(void) changeStateIndicator:(ATSpeechRecognizerState) state{
if(state==ATSpeechRecognizerStateStopped){
//Speech Recognizer is Stopped
_lblState.text = #"Stopped";
}
else{
//Speech Recognizer is running
_lblState.text = #"Running";
}
_txtView.text = #"";
}
-(void) sendErrorInfoToViewController:(NSString *)errorMessage{
[self showPopUpForErrorMessage:errorMessage]; /*Some error occured. Show it to user*/
}
To Start Conversion of Voice to Text:
- (IBAction)btnRecordTapped:(id)sender {
if(!isRecAllowed){
[self showPopUpForErrorMessage:#"Speech recognition is either not authorized or available for this device. Please authorize the operation or upgrade to latest iOS. If you have done all this, check your internet connectivity"];
}
else{
[[ATSpeechRecognizer sharedObject] toggleRecording]; /*If speech Recognizer is running, it will turn it off. if it is off, it will set it on*/
/*
If you want to do it mannually, use startAudioEngine method and stopAudioEngine method to explicitly perform those operations instead of toggleRecording
*/
}
}
And that's it. All the further explanation you need is in code comments. Ping me if you need further explanation.

Answer 2

Here is the full code for the same:
import UIKit
import Speech
public class ViewController: UIViewController, SFSpeechRecognizerDelegate {
// MARK: Properties
private let speechRecognizer = SFSpeechRecognizer(locale: Locale(identifier: "en-US"))!
private var recognitionRequest: SFSpeechAudioBufferRecognitionRequest?
private var recognitionTask: SFSpeechRecognitionTask?
private let audioEngine = AVAudioEngine()
#IBOutlet var textView : UITextView!
#IBOutlet var recordButton : UIButton!
// MARK: UIViewController
public override func viewDidLoad() {
super.viewDidLoad()
// Disable the record buttons until authorization has been granted.
recordButton.isEnabled = false
}
override public func viewDidAppear(_ animated: Bool) {
speechRecognizer.delegate = self
SFSpeechRecognizer.requestAuthorization { authStatus in
/*
The callback may not be called on the main thread. Add an
operation to the main queue to update the record button's state.
*/
OperationQueue.main.addOperation {
switch authStatus {
case .authorized:
self.recordButton.isEnabled = true
case .denied:
self.recordButton.isEnabled = false
self.recordButton.setTitle("User denied access to speech recognition", for: .disabled)
case .restricted:
self.recordButton.isEnabled = false
self.recordButton.setTitle("Speech recognition restricted on this device", for: .disabled)
case .notDetermined:
self.recordButton.isEnabled = false
self.recordButton.setTitle("Speech recognition not yet authorized", for: .disabled)
}
}
}
}
private func startRecording() throws {
// Cancel the previous task if it's running.
if let recognitionTask = recognitionTask {
recognitionTask.cancel()
self.recognitionTask = nil
}
let audioSession = AVAudioSession.sharedInstance()
try audioSession.setCategory(AVAudioSessionCategoryRecord)
try audioSession.setMode(AVAudioSessionModeMeasurement)
try audioSession.setActive(true, with: .notifyOthersOnDeactivation)
recognitionRequest = SFSpeechAudioBufferRecognitionRequest()
guard let inputNode = audioEngine.inputNode else { fatalError("Audio engine has no input node") }
guard let recognitionRequest = recognitionRequest else { fatalError("Unable to created a SFSpeechAudioBufferRecognitionRequest object") }
// Configure request so that results are returned before audio recording is finished
recognitionRequest.shouldReportPartialResults = true
// A recognition task represents a speech recognition session.
// We keep a reference to the task so that it can be cancelled.
recognitionTask = speechRecognizer.recognitionTask(with: recognitionRequest) { result, error in
var isFinal = false
if let result = result {
self.textView.text = result.bestTranscription.formattedString
isFinal = result.isFinal
}
if error != nil || isFinal {
self.audioEngine.stop()
inputNode.removeTap(onBus: 0)
self.recognitionRequest = nil
self.recognitionTask = nil
self.recordButton.isEnabled = true
self.recordButton.setTitle("Start Recording", for: [])
}
}
let recordingFormat = inputNode.outputFormat(forBus: 0)
inputNode.installTap(onBus: 0, bufferSize: 1024, format: recordingFormat) { (buffer: AVAudioPCMBuffer, when: AVAudioTime) in
self.recognitionRequest?.append(buffer)
}
audioEngine.prepare()
try audioEngine.start()
textView.text = "(Go ahead, I'm listening)"
}
// MARK: SFSpeechRecognizerDelegate
public func speechRecognizer(_ speechRecognizer: SFSpeechRecognizer, availabilityDidChange available: Bool) {
if available {
recordButton.isEnabled = true
recordButton.setTitle("Start Recording", for: [])
} else {
recordButton.isEnabled = false
recordButton.setTitle("Recognition not available", for: .disabled)
}
}
// MARK: Interface Builder actions
#IBAction func recordButtonTapped() {
if audioEngine.isRunning {
audioEngine.stop()
recognitionRequest?.endAudio()
recordButton.isEnabled = false
recordButton.setTitle("Stopping", for: .disabled)
} else {
try! startRecording()
recordButton.setTitle("Stop recording", for: [])
}
}
}