gpt4 book ai didi

ios - 如何使用效果音频单元?

转载 作者:行者123 更新时间:2023-11-29 13:08:41 60 4
gpt4 key购买 nike

我正在尝试播放 8000hz 样本并将它们传递给效果器。
我的目标是提高音量(示例代码还没有)。
我想我需要一个效果音频单元,链接到远程音频单元。
我还了解到效果器对其可以处理的格式非常严格,主要需要 44.1khz 样本、浮点和 32 位样本 (.
所以,我添加了一个转换器单元。
此外,由于我认为(虽然不确定)iOS 无法播放 32 位样本(感谢 @hotpaw2!)- 我添加了另一个转换回 16 位的转换。问题是,我在初始化音频图时总是收到错误 -10868。
我也没有最后一个转换单位。
如果我将转换单元连接到输出(无效果单元),一切正常(8k 样本播放正常)。
怎么回事?

    /* Must use play & record category, for reasons beyond the scope of this question */
[[AVAudioSession sharedInstance] setCategory:AVAudioSessionCategoryPlayAndRecord withOptions:AVAudioSessionCategoryOptionDuckOthers | AVAudioSessionCategoryOptionDefaultToSpeaker | AVAudioSessionCategoryOptionAllowBluetooth error:nil];
NSError* err = nil;
if (![[AVAudioSession sharedInstance] setPreferredSampleRate:44100 error:&err]){
NSLog(@"%@",err);
}
AudioUnit effect,convert,output,oconvert;
AUNode neffect,nconvert,noutput,noconvert;
AUGraph graph;
AudioComponentDescription deffect,dconvert,doutput;
AudioStreamBasicDescription in_format,out_format,effect_format;
// Formats
memset(&in_format,0,sizeof(in_format));
memset(&out_format, 0, sizeof(out_format));
memset(&effect_format, 0, sizeof(effect_format));
in_format.mSampleRate = 8000;
in_format.mChannelsPerFrame = 1;
in_format.mFormatID = kAudioFormatLinearPCM;
in_format.mFormatFlags = kLinearPCMFormatFlagIsSignedInteger | kLinearPCMFormatFlagIsPacked;
in_format.mBitsPerChannel = 16;
in_format.mFramesPerPacket = 1;
in_format.mBytesPerFrame = in_format.mChannelsPerFrame * (in_format.mBitsPerChannel / 8);
in_format.mBytesPerPacket = in_format.mBytesPerFrame * in_format.mFramesPerPacket;
out_format.mSampleRate = 44100;
out_format.mChannelsPerFrame = 1;
out_format.mFormatID = kAudioFormatLinearPCM;
out_format.mFormatFlags = kLinearPCMFormatFlagIsSignedInteger | kLinearPCMFormatFlagIsPacked;
out_format.mBitsPerChannel = 16;
out_format.mFramesPerPacket = 1;
out_format.mBytesPerFrame = out_format.mChannelsPerFrame * (out_format.mBitsPerChannel / 8);
out_format.mBytesPerPacket = out_format.mBytesPerFrame * out_format.mFramesPerPacket;
effect_format.mSampleRate = 44100;
effect_format.mChannelsPerFrame = 1;
effect_format.mFormatID = kAudioFormatLinearPCM;
effect_format.mFormatFlags = kAudioFormatFlagsNativeFloatPacked;
effect_format.mBitsPerChannel = 32;
effect_format.mFramesPerPacket = 1;
effect_format.mBytesPerFrame = effect_format.mChannelsPerFrame * (effect_format.mBitsPerChannel / 8);
effect_format.mBytesPerPacket = effect_format.mBytesPerFrame * effect_format.mFramesPerPacket;

// Descriptions
memset(&doutput, 0, sizeof(doutput));
memset(&deffect, 0, sizeof(deffect));
memset(&dconvert, 0, sizeof(dconvert));
doutput.componentType = kAudioUnitType_Output;
doutput.componentSubType = kAudioUnitSubType_RemoteIO;
doutput.componentManufacturer = deffect.componentManufacturer = dconvert.componentManufacturer = kAudioUnitManufacturer_Apple;
dconvert.componentType = kAudioUnitType_FormatConverter;
dconvert.componentSubType = kAudioUnitSubType_AUConverter;
deffect.componentType = kAudioUnitType_Effect;
deffect.componentSubType = kAudioUnitSubType_DynamicsProcessor;
// Create graph
SdCheck(NewAUGraph(&graph));

// Create nodes;
SdCheck(AUGraphAddNode(graph, &deffect, &neffect));
SdCheck(AUGraphAddNode(graph, &doutput, &noutput));
SdCheck(AUGraphAddNode(graph, &dconvert, &nconvert));
SdCheck(AUGraphAddNode(graph, &dconvert, &noconvert));
// Open graph
SdCheck(AUGraphOpen(graph));
// Get units
SdCheck(AUGraphNodeInfo(graph, neffect,NULL, &effect));
SdCheck(AUGraphNodeInfo(graph, noutput,NULL, &output));
SdCheck(AUGraphNodeInfo(graph, nconvert,NULL, &convert));
SdCheck(AUGraphNodeInfo(graph, noconvert,NULL, &oconvert));
// Set formats
SdCheck(AudioUnitSetProperty (output,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Input,
0,
&out_format,
sizeof(out_format)));
SdCheck(AudioUnitSetProperty (convert,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Input,
0,
&effect_format,
sizeof(effect_format)));
SdCheck(AudioUnitSetProperty (convert,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Output,
0,
&in_format,
sizeof(in_format)));
SdCheck(AudioUnitSetProperty (effect,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Input,
0,
&effect_format,
sizeof(effect_format)));
SdCheck(AudioUnitSetProperty (effect,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Output,
0,
&effect_format,
sizeof(effect_format)));
SdCheck(AudioUnitSetProperty (oconvert,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Input,
0,
&out_format,
sizeof(out_format)));
SdCheck(AudioUnitSetProperty (oconvert,
kAudioUnitProperty_StreamFormat,
kAudioUnitScope_Output,
0,
&effect_format,
sizeof(effect_format)));

// Connect nodes
SdCheck(AUGraphConnectNodeInput(graph, nconvert, 0, neffect, 0));
SdCheck(AUGraphConnectNodeInput(graph, neffect, 0, noconvert, 0));
SdCheck(AUGraphConnectNodeInput(graph, noconvert, 0, noutput, 0));
// Set render callback
AURenderCallbackStruct input;
memset(&input, 0, sizeof(input));
input.inputProc = SdInputProc;
input.inputProcRefCon = (__bridge void*)self;
SdCheck(AUGraphSetNodeInputCallback(graph, nconvert, 0, &input));
// Initialize graph
/*** The following fails with error -10868 (unsupported format) ***/
SdCheck(AUGraphInitialize(graph));

最佳答案

iOS 上的大多数效果音频单元都要求在其输入和输出连接上使用 32 位浮点格式。您的示例代码尝试使用 16 位整数 I/O 配置效果器单元,但这是行不通的。

关于ios - 如何使用效果音频单元?,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/17909969/

60 4 0
Copyright 2021 - 2024 cfsdn All Rights Reserved 蜀ICP备2022000587号
广告合作:1813099741@qq.com 6ren.com