iOS原生語音識別使用的正確姿勢

語音識別技術這幾年在移動App上的應用越來越廣，各種第三方語音識別SDK也是層出不窮，例如科大訊飛、百度語音等，同時引入語音識別技術也是一個提高App逼格的好方法:grin:。所以今年的WWDC上，蘋果開放了他的語音識別的API  。有了語音識別技術我們開發者就可以開發出像Siri這樣炫酷的應用。這不，最近我們也要上這個功能，所以我調研一些這方面的東西。本文主要介紹iOS語音識別SDK的用法。

• 首先需要在plist文件中申請語音識別和麥克風使用權限：

  

• 引入頭文件

  #import <Speech/Speech.h>
  #import <AVFoundation/AVFoundation.h>

• 申請權限

  - (void)viewDidAppear:(BOOL)animated{
    [super viewDidAppear:animated];
    [SFSpeechRecognizer requestAuthorization:^(SFSpeechRecognizerAuthorizationStatus status) {
        dispatch_async(dispatch_get_main_queue(), ^{
            switch (status) {
                case SFSpeechRecognizerAuthorizationStatusNotDetermined:
                    self.recordButton.enabled = NO;
                    [self.recordButton setTitle:@"語音識別未授權" forState:UIControlStateDisabled];
                    break;
                case SFSpeechRecognizerAuthorizationStatusDenied:
                    self.recordButton.enabled = NO;
                    [self.recordButton setTitle:@"用戶未授權使用語音識別" forState:UIControlStateDisabled];
                    break;
                case SFSpeechRecognizerAuthorizationStatusRestricted:
                    self.recordButton.enabled = NO;
                    [self.recordButton setTitle:@"語音識別在這臺設備上受到限制" forState:UIControlStateDisabled];
  
                    break;
                case SFSpeechRecognizerAuthorizationStatusAuthorized:
                    self.recordButton.enabled = YES;
                    [self.recordButton setTitle:@"開始錄音" forState:UIControlStateNormal];
                    break;
  
                default:
                    break;
            }
  
        });
    }];
  }

• 核心代碼

  • 識別實時音頻流，如下

    - (void)startRecording{
        if (_recognitionTask) {
           [_recognitionTask cancel];
            _recognitionTask = nil;
         }
    
    AVAudioSession *audioSession = [AVAudioSession sharedInstance];
    NSError *error;
    [audioSession setCategory:AVAudioSessionCategoryRecord error:&error];
    NSParameterAssert(!error);
    [audioSession setMode:AVAudioSessionModeMeasurement error:&error];
    NSParameterAssert(!error);
    [audioSession setActive:YES withOptions:AVAudioSessionSetActiveOptionNotifyOthersOnDeactivation error:&error];
    NSParameterAssert(!error);
    
    _recognitionRequest = [[SFSpeechAudioBufferRecognitionRequest alloc] init];
    AVAudioInputNode *inputNode = self.audioEngine.inputNode;
    NSAssert(inputNode, @"錄入設備沒有準備好");
    NSAssert(_recognitionRequest, @"請求初始化失敗");
    _recognitionRequest.shouldReportPartialResults = YES;
    __weak typeof(self) weakSelf = self;
    _recognitionTask = [self.speechRecognizer recognitionTaskWithRequest:_recognitionRequest resultHandler:^(SFSpeechRecognitionResult * _Nullable result, NSError * _Nullable error) {
        __strong typeof(weakSelf) strongSelf = weakSelf;
        BOOL isFinal = NO;
        if (result) {
            strongSelf.resultStringLable.text = result.bestTranscription.formattedString;
            isFinal = result.isFinal;
        }
        if (error || isFinal) {
            [self.audioEngine stop];
            [inputNode removeTapOnBus:0];
            strongSelf.recognitionTask = nil;
            strongSelf.recognitionRequest = nil;
            strongSelf.recordButton.enabled = YES;
            [strongSelf.recordButton setTitle:@"開始錄音" forState:UIControlStateNormal];
        }
    
    }];
    
    AVAudioFormat *recordingFormat = [inputNode outputFormatForBus:0];
    [inputNode installTapOnBus:0 bufferSize:1024 format:recordingFormat block:^(AVAudioPCMBuffer * _Nonnull buffer, AVAudioTime * _Nonnull when) {
        __strong typeof(weakSelf) strongSelf = weakSelf;
        if (strongSelf.recognitionRequest) {
            [strongSelf.recognitionRequest appendAudioPCMBuffer:buffer];
        }
    }];
    
    [self.audioEngine prepare];
    [self.audioEngine startAndReturnError:&error];
     NSParameterAssert(!error);
    self.resultStringLable.text = @"正在錄音。。。";
    }

    - (SFSpeechRecognizer *)speechRecognizer{
    if (!_speechRecognizer) {
        //要為語音識別對象設置語言，這里設置的是中文
        NSLocale *local =[[NSLocale alloc] initWithLocaleIdentifier:@"zh_CN"];
    
        _speechRecognizer =[[SFSpeechRecognizer alloc] initWithLocale:local];
        _speechRecognizer.delegate = self;
    }
    return _speechRecognizer;
    }

  • 識別本地音頻文件，如下

    - (IBAction)recognizeLocalAudioFile:(UIButton *)sender {   
    NSLocale *local =[[NSLocale alloc] initWithLocaleIdentifier:@"zh_CN"];
    SFSpeechRecognizer *localRecognizer =[[SFSpeechRecognizer alloc] initWithLocale:local];
    NSURL *url =[[NSBundle mainBundle] URLForResource:@"錄音.m4a" withExtension:nil];
    if (!url) return;
    SFSpeechURLRecognitionRequest *res =[[SFSpeechURLRecognitionRequest alloc] initWithURL:url];
    [localRecognizer recognitionTaskWithRequest:res resultHandler:^(SFSpeechRecognitionResult * _Nullable result, NSError * _Nullable error) {
        if (error) {
            NSLog(@"語音識別解析失敗,%@",error);
        }
        else
        {
            self.resultStringLable.text = result.bestTranscription.formattedString;
        }
    }];
    }

• 注意
  • iOS語音識別Api只支持iOS10SDK以及以后的版本，開發工具至少要Xcode8.0。

來自：http://www.jianshu.com/p/a9c64ac2c586