在硬件上运行时没有可用的音频引擎输入通道

Question

我正在尝试在我的音频引擎的语音识别inputNode上设置点击。以下代码在模拟器中运行良好，但当我在运行 iOS 13.1.2 的iPhone X上运行它时则不行。

以下是相关代码：

if (audioEngine == nil || (!audioEngine!.isRunning)) {
print("Setting up audioEngine")
if audioEngine == nil {
audioEngine = AVAudioEngine.init()
}
if(audioEngine!.inputNode.inputFormat(forBus: 0).channelCount == 0){
print("Not enough available inputs!")
}
let recognitionFormat = audioEngine!.inputNode.inputFormat(forBus: 0)
audioEngine!.inputNode.installTap(onBus: 0, bufferSize: 1024, format: recognitionFormat) {
(buffer: AVAudioPCMBuffer, when: AVAudioTime) in
self.recognitionRequest?.append(buffer)
}
audioEngine?.prepare()
try audioEngine?.start()
}

我在模拟器中运行没有错误，语音识别工作正常。但是，在代码上运行时，我得到了以下结果：

Setting up audioEngine
[aurioc] AURemoteIO.cpp:1086:Initialize: failed: -10851 (enable 1, outf< 2 ch,      0 Hz, Float32, non-inter> inf< 2 ch,      0 Hz, Float32, non-inter>)
[aurioc] AURemoteIO.cpp:1086:Initialize: failed: -10851 (enable 1, outf< 2 ch,      0 Hz, Float32, non-inter> inf< 2 ch,      0 Hz, Float32, non-inter>)
Not enough available inputs!

如果我在channelCount == 0后不返回，应用程序将崩溃并显示：

[avae]            AVAEInternal.h:76    required condition is false: [AVAEGraphNode.mm:823:CreateRecordingTap: (IsFormatSampleRateAndChannelCountValid(format))]
*** Terminating app due to uncaught exception 'com.apple.coreaudio.avfaudio', reason: 'required condition is false: IsFormatSampleRateAndChannelCountValid(format)'

我的音频会话设置如下：

try audioSession.setCategory(.playAndRecord, options: [.defaultToSpeaker])
try audioSession.setActive(true)

检查可用输入可得到：

LIST OF INPUTS: [<AVAudioSessionPortDescription: 0x280eac6b0, type = MicrophoneBuiltIn; name = iPhone Microphone; UID = Built-In Microphone; selectedDataSource = Bottom>]

我被难住了。有什么帮助吗？

Answer 1

我想出了问题所在。问题与Flutter有关(这个应用程序正在使用它，但我认为不相关。事实证明，它是相关的。显然，FlutterAppDelegate@application方法没有在主队列上运行，这是我开始AudioSession的地方。当我像这样开始AudioSession时，我的问题消失了：

DispatchQueue.main.async {
let audioSession = AVAudioSession.sharedInstance()
do {
try audioSession.setCategory(.playAndRecord, mode: .spokenAudio, options: .defaultToSpeaker)
try audioSession.setActive(true, options: .notifyOthersOnDeactivation)
} catch {
print("Failed to setup audio session")
}
}