iOS語音書寫功能
編輯：IOS開發綜合
最近在項目開發中，需要將語音識別轉換成文本的功能。研究了下科大訊飛，附上Demo分享給大家。
研發前先得做一些准備。
1、注冊科大訊飛開發者帳號（http://www.xfyun.cn）
2、下載開發平台（iOS、或android，或其他）所需要的SDK（SDK包含：說明文檔、SDK即iflyMSC.framework、Demo）
3、項目中添加SDK（添加時，先將SDK復制粘貼到項目文件，再通過addframe的方法添加到項目引用），及相關聯的framework
添加方法：TARGETS-Build Phases-Link Binary With Libraries-"+"-Choose frameworks and libraries to add-add other，或選擇對應的framework-add
4、使用時要添加對應的頭文件
特別說明：
1、使用SDK關聯的APPID存在於下載的Demo中，如果SDK有替換的話APPID應該跟著一起替換。
2、在使用前，務必在AppDelegate的方法中"
- (BOOL)application:(UIApplication *)application didFinishLaunchingWithOptions:(NSDictionary *)launchOptions {}"進行初始化操作。
3、需要有網絡的情況下才能使用。
如圖
下載的科大訊飛SDK文件
Demo中的APPID
添加SDK
添加關聯framework
語音轉文件實現代碼
.h文件 #import// 導入頭文件 #import "iflyMSC.framework/Headers/IFlyMSC.h" #import "iflyMSC.framework/Headers/IFlySpeechUtility.h" #import "iflyMSC/IFlySpeechConstant.h" #pragma mark - 初始化參數類 /**************************************************************************/ @interface IATConfig : NSObject + (IATConfig *)sharedInstance; + (NSString *)mandarin; + (NSString *)cantonese; + (NSString *)henanese; + (NSString *)chinese; + (NSString *)english; + (NSString *)lowSampleRate; + (NSString *)highSampleRate; + (NSString *)isDot; + (NSString *)noDot; /** 以下參數，需要通過 iFlySpeechRecgonizer 進行設置 */ @property (nonatomic, strong) NSString *speechTimeout; @property (nonatomic, strong) NSString *vadEos; @property (nonatomic, strong) NSString *vadBos; @property (nonatomic, strong) NSString *language; @property (nonatomic, strong) NSString *accent; @property (nonatomic, strong) NSString *dot; @property (nonatomic, strong) NSString *sampleRate; /** 以下參數無需設置 不必關 */ @property (nonatomic, assign) BOOL haveView; @property (nonatomic, strong) NSArray *accentIdentifer; @property (nonatomic, strong) NSArray *accentNickName; @end /**************************************************************************/ #pragma mark - 語音聽寫類 @interface VoiceConversion : NSObject /// 啟動初始化語音程序 + (void)VoiceInitialize; /// 開始錄音 - (void)voiceStart:(void (^)(BOOL isStart))startListening speechBegin:(void (^)(void))begin speechEnd:(void (^)(void))end speechError:(void (^)(BOOL isSuccess))error speechResult:(void (^)(NSString *text))result speechVolume:(void (^)(int volume))volume; /// 取消錄音 - (void)voiceCancel; /// 停止錄音 - (void)voiceStop; @end
.m文件
#import "VoiceConversion.h"
#pragma mark - 初始化參數類
/**************************************************************************/
static NSString *const PUTONGHUA = @"mandarin";
static NSString *const YUEYU = @"cantonese";
static NSString *const HENANHUA = @"henanese";
static NSString *const ENGLISH = @"en_us";
static NSString *const CHINESE = @"zh_cn";
@implementation IATConfig
- (id)init
{
self = [super init];
if (self)
{
[self defaultSetting];
return self;
}
return nil;
}
+ (IATConfig *)sharedInstance
{
static IATConfig * instance = nil;
static dispatch_once_t predict;
dispatch_once(&predict, ^{
instance = [[IATConfig alloc] init];
});
return instance;
}
- (void)defaultSetting
{
_speechTimeout = @"30000";
_vadEos = @"3000";
_vadBos = @"3000";
_dot = @"1";
_sampleRate = @"16000";
_language = CHINESE;
_accent = PUTONGHUA;
_haveView = NO;//默認是不dai界面的
_accentNickName = [[NSArray alloc] initWithObjects:@"粵語", @"普通話", @"河南話", @"英文", nil];
}
+ (NSString *)mandarin
{
return PUTONGHUA;
}
+ (NSString *)cantonese
{
return YUEYU;
}
+ (NSString *)henanese
{
return HENANHUA;
}
+ (NSString *)chinese
{
return CHINESE;
}
+ (NSString *)english
{
return ENGLISH;
}
+ (NSString *)lowSampleRate
{
return @"8000";
}
+ (NSString *)highSampleRate
{
return @"16000";
}
+ (NSString *)isDot
{
return @"1";
}
+ (NSString *)noDot
{
return @"0";
}
@end
/**************************************************************************/
#pragma mark - 語音聽寫類
static NSString *const VoiceAPPID = @"572016e4";
static NSString *const VoiceTimeOut = @"20000";
@interface VoiceConversion ()
@property (nonatomic, strong) NSMutableString *resultText;
@property (nonatomic, strong) IFlySpeechRecognizer *iFlySpeechRecognizer;
@property (nonatomic, copy) void (^beginSpeech)(void);
@property (nonatomic, copy) void (^endSpeech)(void);
@property (nonatomic, copy) void (^errorSpeech)(BOOL isSuccess);
@property (nonatomic, copy) void (^resultSpeech)(NSString *text);
@property (nonatomic, copy) void (^volumeSpeech)(int volume);
@end
@implementation VoiceConversion
#pragma mark 初始化------------
/// 啟動初始化語音程序
+ (void)VoiceInitialize
{
// 設置sdk的log等級，log保存在下面設置的工作路徑中
[IFlySetting setLogFile:LVL_ALL];
// 打開輸出在console的log開關
[IFlySetting showLogcat:YES];
// 設置sdk的工作路徑
NSArray *paths = NSSearchPathForDirectoriesInDomains(NSCachesDirectory, NSUserDomainMask, YES);
NSString *cachePath = [paths objectAtIndex:0];
[IFlySetting setLogFilePath:cachePath];
// Appid是應用的身份信息,具有唯一性,初始化時必須要傳入Appid。初始化是一個異步過程,可放在 App 啟動時執行初始化,具體代碼可以參 照 Demo 的 MSCAppDelegate.m。未初始化時使用服務,一般會返回錯誤碼 10111.
NSString *initString = [[NSString alloc] initWithFormat:@"appid=%@", VoiceAPPID];
[IFlySpeechUtility createUtility:initString];
}
#pragma mark 實例化------------
- (void)dealloc
{
[self voiceCancel];
}
- (NSMutableString *)resultText
{
if (!_resultText)
{
_resultText = [[NSMutableString alloc] init];
}
return _resultText;
}
- (IFlySpeechRecognizer *)iFlySpeechRecognizer
{
if (_iFlySpeechRecognizer == nil)
{
_iFlySpeechRecognizer = [IFlySpeechRecognizer sharedInstance];
[_iFlySpeechRecognizer setParameter:@"" forKey:[IFlySpeechConstant PARAMS]];
// 設置聽寫模式
[_iFlySpeechRecognizer setParameter:@"iat" forKey:[IFlySpeechConstant IFLY_DOMAIN]];
}
return _iFlySpeechRecognizer;
}
- (void)initializeVoice
{
self.iFlySpeechRecognizer.delegate = self;
IATConfig *instance = [IATConfig sharedInstance];
// 設置最長錄音時間
[self.iFlySpeechRecognizer setParameter:instance.speechTimeout forKey:[IFlySpeechConstant SPEECH_TIMEOUT]];
// 設置後端點
[self.iFlySpeechRecognizer setParameter:instance.vadEos forKey:[IFlySpeechConstant VAD_EOS]];
// 設置前端點
[self.iFlySpeechRecognizer setParameter:instance.vadBos forKey:[IFlySpeechConstant VAD_BOS]];
// 網絡等待時間
[self.iFlySpeechRecognizer setParameter:@"20000" forKey:[IFlySpeechConstant NET_TIMEOUT]];
// 設置采樣率，推薦使用16K
[self.iFlySpeechRecognizer setParameter:instance.sampleRate forKey:[IFlySpeechConstant SAMPLE_RATE]];
if ([instance.language isEqualToString:[IATConfig chinese]])
{
// 設置語言
[self.iFlySpeechRecognizer setParameter:instance.language forKey:[IFlySpeechConstant LANGUAGE]];
// 設置方言
[self.iFlySpeechRecognizer setParameter:instance.accent forKey:[IFlySpeechConstant ACCENT]];
}
else if ([instance.language isEqualToString:[IATConfig english]])
{
[self.iFlySpeechRecognizer setParameter:instance.language forKey:[IFlySpeechConstant LANGUAGE]];
}
// 設置是否返回標點符號
[self.iFlySpeechRecognizer setParameter:instance.dot forKey:[IFlySpeechConstant ASR_PTT]];
}
#pragma mark 語音聽寫方法------------
/// 開始錄音
- (void)voiceStart:(void (^)(BOOL isStart))startListening speechBegin:(void (^)(void))begin speechEnd:(void (^)(void))end speechError:(void (^)(BOOL isSuccess))error speechResult:(void (^)(NSString *text))result speechVolume:(void (^)(int volume))volume
{
[self.resultText setString:@""];
// 回調設置
self.beginSpeech = [begin copy];
self.endSpeech = [end copy];
self.errorSpeech = [error copy];
self.resultSpeech = [result copy];
self.volumeSpeech = [volume copy];
// 初始化設置
[self initializeVoice];
[self.iFlySpeechRecognizer cancel];
// 設置音頻來源為麥克風
[self.iFlySpeechRecognizer setParameter:IFLY_AUDIO_SOURCE_MIC forKey:@"audio_source"];
// 設置聽寫結果格式為json
[self.iFlySpeechRecognizer setParameter:@"json" forKey:[IFlySpeechConstant RESULT_TYPE]];
// 保存錄音文件，保存在sdk工作路徑中，如未設置工作路徑，則默認保存在library/cache下
[self.iFlySpeechRecognizer setParameter:@"asr.pcm" forKey:[IFlySpeechConstant ASR_AUDIO_PATH]];
BOOL isStart = [self.iFlySpeechRecognizer startListening];
if (startListening)
{
// 如果開始錄音失敗，可能是上次請求未結束，暫不支持多路並發
startListening(isStart);
}
}
/// 取消聽寫
- (void)voiceCancel
{
[self.iFlySpeechRecognizer cancel];
}
/// 停止錄音
- (void)voiceStop
{
[self.iFlySpeechRecognizer stopListening];
}
#pragma mark IFlySpeechRecognizerDelegate------------
/**
識別結果返回代理
@param :results識別結果
@ param :isLast 表示是否最後一次結果
*/
- (void)onResults:(NSArray *)results isLast:(BOOL)isLast
{
NSMutableString *resultString = [[NSMutableString alloc] init];
NSDictionary *dic = results[0];
for (NSString *key in dic)
{
[resultString appendFormat:@"%@",key];
}
NSString *resultFromJson = [[self class] stringFromJson:resultString];
NSString *resultTextTemp = [NSString stringWithFormat:@"%@%@", self.resultText, resultFromJson];
[self.resultText setString:resultTextTemp];
if (self.resultSpeech)
{
self.resultSpeech(self.resultText);
}
}
/**
識別會話結束返回代理
@ param error 錯誤碼,error.errorCode=0表示正常結束,非0表示發生錯誤。
*/
- (void)onError:(IFlySpeechError *)error
{
if (self.errorSpeech)
{
BOOL isSuccess = (0 == error.errorCode);
self.errorSpeech(isSuccess);
}
}
/**
停止錄音回調
*/
- (void)onEndOfSpeech
{
if (self.endSpeech)
{
self.endSpeech();
}
}
/**
開始識別回調
*/
- (void)onBeginOfSpeech
{
if (self.beginSpeech)
{
self.beginSpeech();
}
}
/**
音量回調函數 volume 0-30
*/
- (void)onVolumeChanged:(int)volume
{
if (self.volumeSpeech)
{
self.volumeSpeech(volume);
}
}
#pragma mark 解析方法------------
/**************************************************************************/
/**
解析命令詞返回的結果
*/
+ (NSString *)stringFromAsr:(NSString *)params;
{
NSMutableString * resultString = [[NSMutableString alloc] init];
NSString *inputString = nil;
NSArray *array = [params componentsSeparatedByString:@"\n"];
for (int index = 0; index < array.count; index++)
{
NSRange range;
NSString *line = [array objectAtIndex:index];
NSRange idRange = [line rangeOfString:@"id="];
NSRange nameRange = [line rangeOfString:@"name="];
NSRange confidenceRange = [line rangeOfString:@"confidence="];
NSRange grammarRange = [line rangeOfString:@" grammar="];
NSRange inputRange = [line rangeOfString:@"input="];
if (confidenceRange.length == 0 || grammarRange.length == 0 || inputRange.length == 0 )
{
continue;
}
// check nomatch
if (idRange.length != 0)
{
NSUInteger idPosX = idRange.location + idRange.length;
NSUInteger idLength = nameRange.location - idPosX;
range = NSMakeRange(idPosX, idLength);
NSString *subString = [line substringWithRange:range];
NSCharacterSet *subSet = [NSCharacterSet whitespaceAndNewlineCharacterSet];
NSString *idValue = [subString stringByTrimmingCharactersInSet:subSet];
if ([idValue isEqualToString:@"nomatch"])
{
return @"";
}
}
// Get Confidence Value
NSUInteger confidencePosX = confidenceRange.location + confidenceRange.length;
NSUInteger confidenceLength = grammarRange.location - confidencePosX;
range = NSMakeRange(confidencePosX,confidenceLength);
NSString *score = [line substringWithRange:range];
NSUInteger inputStringPosX = inputRange.location + inputRange.length;
NSUInteger inputStringLength = line.length - inputStringPosX;
range = NSMakeRange(inputStringPosX , inputStringLength);
inputString = [line substringWithRange:range];
[resultString appendFormat:@"%@ 置信度%@\n",inputString, score];
}
return resultString;
}
/**
解析聽寫json格式的數據
params例如：
{"sn":1,"ls":true,"bg":0,"ed":0,"ws":[{"bg":0,"cw":[{"w":"白日","sc":0}]},{"bg":0,"cw":[{"w":"依山","sc":0}]},{"bg":0,"cw":[{"w":"盡","sc":0}]},{"bg":0,"cw":[{"w":"黃河入海流","sc":0}]},{"bg":0,"cw":[{"w":"。","sc":0}]}]}
*/
+ (NSString *)stringFromJson:(NSString *)params
{
if (params == NULL)
{
return nil;
}
NSMutableString *tempStr = [[NSMutableString alloc] init];
// 返回的格式必須為utf8的,否則發生未知錯誤
NSData *dataJSON = [params dataUsingEncoding:NSUTF8StringEncoding];
NSDictionary *resultDic = [NSJSONSerialization JSONObjectWithData:dataJSON options:kNilOptions error:nil];
if (resultDic != nil)
{
NSArray *wordArray = [resultDic objectForKey:@"ws"];
for (int i = 0; i < [wordArray count]; i++)
{
NSDictionary *wsDic = [wordArray objectAtIndex:i];
NSArray *cwArray = [wsDic objectForKey:@"cw"];
for (int j = 0; j < [cwArray count]; j++)
{
NSDictionary *wDic = [cwArray objectAtIndex:j];
NSString *str = [wDic objectForKey:@"w"];
[tempStr appendString: str];
}
}
}
return tempStr;
}
/**
解析語法識別返回的結果
*/
+ (NSString *)stringFromABNFJson:(NSString *)params
{
if (params == NULL)
{
return nil;
}
NSMutableString *tempStr = [[NSMutableString alloc] init];
NSData *dataJSON = [params dataUsingEncoding:NSUTF8StringEncoding];
NSDictionary *resultDic = [NSJSONSerialization JSONObjectWithData:dataJSON options:kNilOptions error:nil];
NSArray *wordArray = [resultDic objectForKey:@"ws"];
for (int i = 0; i < [wordArray count]; i++)
{
NSDictionary *wsDic = [wordArray objectAtIndex:i];
NSArray *cwArray = [wsDic objectForKey:@"cw"];
for (int j = 0; j < [cwArray count]; j++)
{
NSDictionary *wDic = [cwArray objectAtIndex:j];
NSString *str = [wDic objectForKey:@"w"];
NSString *score = [wDic objectForKey:@"sc"];
[tempStr appendString: str];
[tempStr appendFormat:@" 置信度:%@",score];
[tempStr appendString: @"\n"];
}
}
return tempStr;
}
/**************************************************************************/
@end
使用
初始化方法
/// 啟動初始化語音程序
+ (void)VoiceInitialize
{
// 設置sdk的log等級，log保存在下面設置的工作路徑中
[IFlySetting setLogFile:LVL_ALL];
// 打開輸出在console的log開關
[IFlySetting showLogcat:YES];
// 設置sdk的工作路徑
NSArray *paths = NSSearchPathForDirectoriesInDomains(NSCachesDirectory, NSUserDomainMask, YES);
NSString *cachePath = [paths objectAtIndex:0];
[IFlySetting setLogFilePath:cachePath];
// Appid是應用的身份信息,具有唯一性,初始化時必須要傳入Appid。初始化是一個異步過程,可放在 App 啟動時執行初始化,具體代碼可以參 照 Demo 的 MSCAppDelegate.m。未初始化時使用服務,一般會返回錯誤碼 10111.
NSString *initString = [[NSString alloc] initWithFormat:@"appid=%@", VoiceAPPID];
[IFlySpeechUtility createUtility:initString];
}
初始化調用
- (BOOL)application:(UIApplication *)application didFinishLaunchingWithOptions:(NSDictionary *)launchOptions
{
// Override point for customization after application launch.
[VoiceConversion VoiceInitialize];
return YES;
}
#import "VoiceConversion.h"
@interface ViewController ()
@property (nonatomic, strong) VoiceConversion *voiceConversion;
@property (nonatomic, strong) UILabel *messageLabel;
@end
@implementation ViewController
- (void)viewDidLoad {
[super viewDidLoad];
// Do any additional setup after loading the view, typically from a nib.
UIBarButtonItem *startItem = [[UIBarButtonItem alloc] initWithTitle:@"start" style:UIBarButtonItemStyleDone target:self action:@selector(startItemClick:)];
UIBarButtonItem *stopItem = [[UIBarButtonItem alloc] initWithTitle:@"stop" style:UIBarButtonItemStyleDone target:self action:@selector(stopItemClick:)];
UIBarButtonItem *cancelItem = [[UIBarButtonItem alloc] initWithTitle:@"cancel" style:UIBarButtonItemStyleDone target:self action:@selector(cancelItemClick:)];
self.navigationItem.rightBarButtonItems = @[startItem, stopItem, cancelItem];
self.title = @"科大訊飛語音";
[self setUI];
}
- (void)didReceiveMemoryWarning {
[super didReceiveMemoryWarning];
// Dispose of any resources that can be recreated.
}
#pragma mark - 視圖
- (void)setUI
{
if ([self respondsToSelector:@selector(setEdgesForExtendedLayout:)])
{
[self setEdgesForExtendedLayout:UIRectEdgeNone];
}
self.messageLabel = [[UILabel alloc] initWithFrame:CGRectMake(10.0, 10.0, CGRectGetWidth(self.view.bounds) - 10.0 * 2, 40.0)];
[self.view addSubview:self.messageLabel];
self.messageLabel.backgroundColor = [UIColor colorWithWhite:0.5 alpha:0.3];
self.messageLabel.textAlignment = NSTextAlignmentCenter;
}
#pragma mark - 響應
- (void)startItemClick:(UIBarButtonItem *)item
{
ViewController __weak *weakSelf = self;
[self.voiceConversion voiceStart:^(BOOL isStart) {
NSLog(@"1 start");
if (isStart)
{
weakSelf.messageLabel.text = @"正在錄音";
}
else
{
weakSelf.messageLabel.text = @"啟動識別服務失敗，請稍後重試";
}
} speechBegin:^{
NSLog(@"2 begin");
} speechEnd:^{
NSLog(@"3 end");
} speechError:^(BOOL isSuccess) {
NSLog(@"4 error");
} speechResult:^(NSString *text) {
NSLog(@"5 result");
weakSelf.messageLabel.text = text;
} speechVolume:^(int volume) {
NSLog(@"6 volume");
NSString *volumeString = [NSString stringWithFormat:@"音量：%d", volume];
weakSelf.messageLabel.text = volumeString;
}];
}
- (void)stopItemClick:(UIBarButtonItem *)item
{
[self.voiceConversion voiceStop];
self.messageLabel.text = @"停止錄音";
}
- (void)cancelItemClick:(UIBarButtonItem *)item
{
[self.voiceConversion voiceCancel];
self.messageLabel.text = @"取消識別";
}
#pragma mark - getter
- (VoiceConversion *)voiceConversion
{
if (!_voiceConversion)
{
_voiceConversion = [[VoiceConversion alloc] init];
}
return _voiceConversion;
}
@end
- 上一頁:Xcode7最新app打包發布詳細過程（二）
- 下一頁:iOS動畫詳解
相關文章
+