我正在使用这个实时音高检测程序： https://github.com/fotock/PitchDetectorExample/tree/1c68491f9c9bff2e851f5711c47e1efe4092f4de
为了我的目的，它运作良好;它有一个频率标签，当你唱一个音高时，标签会记录一个频率，当你唱一个更高的音高时，频率标签会增加。
问题是当我没有在麦克风上唱歌时，频率标签仍会记录一个频率，通常在70Hz左右，但有时即使麦克风关闭也会跳到200Hz。
当音量/数据库足够响亮时，有没有办法让麦克风仅打开？仅在麦克风接收预设幅度时触发的事件侦听器。基本上，我需要一个音频门，如果DB很低，只有线路噪声，那么话筒就会关闭。
以下是来自上述应用程序的音高检测代码 - 未删节。我试图无效，将vDSP代码添加到此代码并读取输入频率的幅度以打开和关闭麦克风。
PitchDetector.m
#import "PitchDetector.h"
#import <Accelerate/Accelerate.h>
#define PD_SYSTEM_VERSION_GREATER_THAN_OR_EQUAL_TO(v) ([[[UIDevice currentDevice] systemVersion] compare:v options:NSNumericSearch] != NSOrderedAscending)
@implementation PitchDetector
@synthesize lowBoundFrequency, hiBoundFrequency, sampleRate, delegate, running;
#pragma mark Initialize Methods
-(id) initWithSampleRate: (float) rate andDelegate: (id<PitchDetectorDelegate>) initDelegate {
return [self initWithSampleRate:rate lowBoundFreq:40 hiBoundFreq:4500 andDelegate:initDelegate];
}
-(id) initWithSampleRate: (float) rate lowBoundFreq: (int) low hiBoundFreq: (int) hi andDelegate: (id<PitchDetectorDelegate>) initDelegate {
self.lowBoundFrequency = low;
self.hiBoundFrequency = hi;
self.sampleRate = rate;
self.delegate = initDelegate;
bufferLength = self.sampleRate/self.lowBoundFrequency;
hann = (float*) malloc(sizeof(float)*bufferLength);
vDSP_hann_window(hann, bufferLength, vDSP_HANN_NORM);
sampleBuffer = (SInt16*) malloc(512);
samplesInSampleBuffer = 0;
result = (float*) malloc(sizeof(float)*bufferLength);
return self;
}
#pragma mark Insert Samples
- (void) addSamples:(SInt16 *)samples inNumberFrames:(int)frames {
int newLength = frames;
if(samplesInSampleBuffer>0) {
newLength += samplesInSampleBuffer;
}
SInt16 *newBuffer = (SInt16*) malloc(sizeof(SInt16)*newLength);
memcpy(newBuffer, sampleBuffer, samplesInSampleBuffer*sizeof(SInt16));
memcpy(&newBuffer[samplesInSampleBuffer], samples, frames*sizeof(SInt16));
free(sampleBuffer);
sampleBuffer = newBuffer;
samplesInSampleBuffer = newLength;
if(samplesInSampleBuffer>(self.sampleRate/self.lowBoundFrequency)) {
if(!self.running) {
[self performSelectorInBackground:@selector(performWithNumFrames:) withObject:[NSNumber numberWithInt:newLength]];
self.running = YES;
}
samplesInSampleBuffer = 0;
} else {
//printf("NOT ENOUGH SAMPLES: %d\n", newLength);
}
}
#pragma mark Perform Auto Correlation
-(void) performWithNumFrames: (NSNumber*) numFrames;
{
int n = numFrames.intValue;
float freq = 0;
SInt16 *samples;
if (PD_SYSTEM_VERSION_GREATER_THAN_OR_EQUAL_TO(@"7.1")) {
@synchronized(self) {
samples = malloc(sizeof(SInt16)*numFrames.intValue);
memcpy(&samples, &sampleBuffer, sizeof(samples));
}
} else {
samples = sampleBuffer;
}
int returnIndex = 0;
float sum;
bool goingUp = false;
float normalize = 0;
for(int i = 0; i<n; i++) {
sum = 0;
for(int j = 0; j<n; j++) {
sum += (samples[j]*samples[j+i])*hann[j];
}
if(i ==0 ) normalize = sum;
result[i] = sum/normalize;
}
for(int i = 0; i<n-8; i++) {
if(result[i]<0) {
i+=2; // no peaks below 0, skip forward at a faster rate
} else {
if(result[i]>result[i-1] && goingUp == false && i >1) {
//local min at i-1
goingUp = true;
} else if(goingUp == true && result[i]<result[i-1]) {
//local max at i-1
if(returnIndex==0 && result[i-1]>result[0]*0.95) {
returnIndex = i-1;
break;
//############### NOTE ##################################
// My implemenation breaks out of this loop when it finds the first peak.
// This is (probably) the greatest source of error, so if you would like to
// improve this algorithm, start here. the next else if() will trigger on
// future local maxima (if you first take out the break; above this paragraph)
//#######################################################
} else if(result[i-1]>result[0]*0.85) {
}
goingUp = false;
}
}
}
freq =self.sampleRate/interp(result[returnIndex-1], result[returnIndex], result[returnIndex+1], returnIndex);
if(freq >= self.lowBoundFrequency && freq <= self.hiBoundFrequency) {
dispatch_async(dispatch_get_main_queue(), ^{
[delegate updatedPitch:freq];
});
}
self.running = NO;
}
float interp(float y1, float y2, float y3, int k);
float interp(float y1, float y2, float y3, int k) {
float d, kp;
d = (y3 - y1) / (2 * (2 * y2 - y1 - y3));
//printf("%f = %d + %f\n", k+d, k, d);
kp = k + d;
return kp;
}
@end
这是初学化麦克风的AudioControll.m类 - 未删节。看看根据数据库打开和关闭麦克风的其他例子，我尝试了许多框架，如音频队列服务和加速框架，但无济于事。
AudioControll.m
#import "AudioController.h"
#define kOutputBus 0
#define kInputBus 1
@implementation AudioController
@synthesize rioUnit, audioFormat, delegate;
+ (AudioController *) sharedAudioManager
{
static AudioController *sharedAudioManager;
@synchronized(self)
{
if (!sharedAudioManager) {
sharedAudioManager = [[AudioController alloc] init];
[sharedAudioManager startAudio];
}
return sharedAudioManager;
}
}
void checkStatus(OSStatus status);
void checkStatus(OSStatus status) {
if(status!=0)
printf("Error: %ld\n", status);
}
#pragma mark init
- (id)init
{
OSStatus status;
status = AudioSessionInitialize(NULL, NULL, NULL, (__bridge void*) self);
checkStatus(status);
// Describe audio component
AudioComponentDescription desc;
desc.componentType = kAudioUnitType_Output;
desc.componentSubType = kAudioUnitSubType_RemoteIO;
desc.componentFlags = 0;
desc.componentFlagsMask = 0;
desc.componentManufacturer = kAudioUnitManufacturer_Apple;
// Get component
AudioComponent inputComponent = AudioComponentFindNext(NULL, &desc);
// Get audio units
status = AudioComponentInstanceNew(inputComponent, &rioUnit);
checkStatus(status);
// Enable IO for recording
UInt32 flag = 1;
status = AudioUnitSetProperty(rioUnit,
kAudioOutputUnitProperty_EnableIO,
kAudioUnitScope_Input,
kInputBus,
&flag,
sizeof(flag));
checkStatus(status);
// Describe format
audioFormat.mSampleRate= 44100.0;
audioFormat.mFormatID= kAudioFormatLinearPCM;
audioFormat.mFormatFlags= kAudioFormatFlagIsSignedInteger | kAudioFormatFlagIsPacked;
audioFormat.mFramesPerPacket= 1;
audioFormat.mChannelsPerFrame= 1;
audioFormat.mBitsPerChannel= 16;
audioFormat.mBytesPerPacket= 2;
audioFormat.mBytesPerFrame= 2;
// Apply format
status = AudioUnitSetProperty(rioUnit,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Output,
kInputBus,
&audioFormat,
sizeof(audioFormat));
checkStatus(status);
status = AudioUnitSetProperty(rioUnit,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Input,
kOutputBus,
&audioFormat,
sizeof(audioFormat));
checkStatus(status);
// Set input callback
AURenderCallbackStruct callbackStruct;
callbackStruct.inputProc = recordingCallback;
callbackStruct.inputProcRefCon = (__bridge void*)self;
status = AudioUnitSetProperty(rioUnit,
kAudioOutputUnitProperty_SetInputCallback,
kAudioUnitScope_Global,
kInputBus,
&callbackStruct,
sizeof(callbackStruct));
checkStatus(status);
// Disable buffer allocation for the recorder
flag = 0;
status = AudioUnitSetProperty(rioUnit, kAudioUnitProperty_ShouldAllocateBuffer, kAudioUnitScope_Global, kInputBus, &flag, sizeof(flag));
// Initialise
UInt32 category = kAudioSessionCategory_PlayAndRecord;
status = AudioSessionSetProperty(kAudioSessionProperty_AudioCategory, sizeof(category), &category);
checkStatus(status);
status = 0;
status = AudioSessionSetActive(YES);
checkStatus(status);
status = AudioUnitInitialize(rioUnit);
checkStatus(status);
return self;
}
#pragma mark Recording Callback
static OSStatus recordingCallback(void *inRefCon,
AudioUnitRenderActionFlags *ioActionFlags,
const AudioTimeStamp *inTimeStamp,
UInt32 inBusNumber,
UInt32 inNumberFrames,
AudioBufferList *ioData) {
AudioController *THIS = (__bridge AudioController*) inRefCon;
THIS->bufferList.mNumberBuffers = 1;
THIS->bufferList.mBuffers[0].mDataByteSize = sizeof(SInt16)*inNumberFrames;
THIS->bufferList.mBuffers[0].mNumberChannels = 1;
THIS->bufferList.mBuffers[0].mData = (SInt16*) malloc(sizeof(SInt16)*inNumberFrames);
OSStatus status;
status = AudioUnitRender(THIS->rioUnit,
ioActionFlags,
inTimeStamp,
inBusNumber,
inNumberFrames,
&(THIS->bufferList));
checkStatus(status);
dispatch_async(dispatch_get_main_queue(), ^{
[THIS.delegate receivedAudioSamples:(SInt16*)THIS->bufferList.mBuffers[0].mData length:inNumberFrames];
});
return noErr;
}
-(void) startAudio
{
OSStatus status = AudioOutputUnitStart(rioUnit);
checkStatus(status);
printf("Audio Initialized - sampleRate: %f\n", audioFormat.mSampleRate);
}
@end