swift AVAudioEngine 和 AVAudioSinkNode sampleRate 转换

Question

一段时间以来我一直遇到这个问题，并编写了以下可以作为应用程序的主视图 Controller 文件运行的 swift 文件。执行后，它将播放一段短促的 1kHz 正弦波。它将同时从音频接口(interface)的输入进行记录。

目前我已经将输出插入到输入中进行测试。但这也可能是内置扬声器和内置麦克风的计算机(运行应用程序前只需检查系统设置中的音量，因为它会自动播放)

我无法得到准确的结果:

import UIKit
import AVFoundation

var globalSampleRate = 48000


class ViewController: UIViewController {
    var micBuffer:[Float] = Array(repeating:0, count:10000)
    var referenceBuffer:[Float] = Array(repeating:0, count:10000)
    var running:Bool = false
    var engine = AVAudioEngine()

    override func viewDidLoad() {
        super.viewDidLoad()

        let syncQueue = DispatchQueue(label:"Audio Engine")
        syncQueue.sync{
            initializeAudioEngine()
            while running == true {
            }
            engine.stop()
            writetoFile(buff: micBuffer, name: "Mic Input")
            writetoFile(buff: referenceBuffer, name: "Reference")

        }
    }

    func initializeAudioEngine(){

        var micBufferPosition:Int = 0
        var refBufferPosition:Int = 0
        let frequency:Float = 1000.0
        let amplitude:Float = 1.0
        let signal = { (time: Float) -> Float in
            return amplitude * sin(2.0 * Float.pi * frequency * time)
        }

        let deltaTime = 1.0 / Float(globalSampleRate)
        var time: Float = 0

        let micSinkNode = AVAudioSinkNode() { (timeStamp, frames, audioBufferList) ->
          OSStatus in

            let ptr = audioBufferList.pointee.mBuffers.mData?.assumingMemoryBound(to: Float.self)
            var monoSamples = [Float]()
            monoSamples.append(contentsOf: UnsafeBufferPointer(start: ptr, count: Int(frames)))
            for frame in 0..<frames {
              self.micBuffer[micBufferPosition + Int(frame)] = monoSamples[Int(frame)]
            }
            micBufferPosition += Int(frames)

            if micBufferPosition > 8000 {
                self.running = false
            }

            return noErr
        }


        let srcNode = AVAudioSourceNode { _, _, frameCount, audioBufferList -> OSStatus in
            let ablPointer = UnsafeMutableAudioBufferListPointer(audioBufferList)
            for frame in 0..<Int(frameCount) {
                let value = signal(time)
                time += deltaTime
                for buffer in ablPointer {
                    let buf: UnsafeMutableBufferPointer<Float> = UnsafeMutableBufferPointer(buffer)
                    buf[frame] = value
                    self.referenceBuffer[refBufferPosition + frame] = value
                }

            }
            refBufferPosition += Int(frameCount)
            return noErr
        }

        let inputFormat = engine.inputNode.inputFormat(forBus: 0)
        let outputFormat = engine.outputNode.outputFormat(forBus: 0)
        let nativeFormat = AVAudioFormat(commonFormat: .pcmFormatFloat32,
            sampleRate: Double(globalSampleRate),
            channels: 1,
            interleaved: false)

        let formatMixer  = AVAudioMixerNode()
        engine.attach(formatMixer)
        engine.attach(micSinkNode)
        engine.attach(srcNode)
        //engine.connect(engine.inputNode, to: micSinkNode, format: inputFormat)
        engine.connect(engine.inputNode, to: formatMixer, format: inputFormat)
        engine.connect(formatMixer, to: micSinkNode, format: nativeFormat)
        engine.connect(srcNode, to: engine.mainMixerNode, format: nativeFormat)
        engine.connect(engine.mainMixerNode, to: engine.outputNode, format: outputFormat)
        print("micSinkNode Format is \(micSinkNode.inputFormat(forBus: 0))")
        print("inputNode Format is \(engine.inputNode.inputFormat(forBus: 0))")
        print("outputNode Format is \(engine.outputNode.outputFormat(forBus: 0))")
        print("formatMixer Format is \(formatMixer.outputFormat(forBus: 0))")

        engine.prepare()
        running = true
        do {
            try engine.start()
        } catch {
            print("Error")
        }
    }

}


func writetoFile(buff:[Float], name:String){

    let outputFormatSettings = [
        AVFormatIDKey:kAudioFormatLinearPCM,
        AVLinearPCMBitDepthKey:32,
        AVLinearPCMIsFloatKey: true,
        AVLinearPCMIsBigEndianKey: true,
        AVSampleRateKey: globalSampleRate,
        AVNumberOfChannelsKey: 1
        ] as [String : Any]

    let fileName = name
    let DocumentDirURL = try! FileManager.default.url(for: .documentDirectory, in: .userDomainMask, appropriateFor: nil, create: true)


    let url = DocumentDirURL.appendingPathComponent(fileName).appendingPathExtension("wav")
    print("FilePath: \(url.path)")

    let audioFile = try? AVAudioFile(forWriting: url, settings: outputFormatSettings, commonFormat: AVAudioCommonFormat.pcmFormatFloat32, interleaved: false)

    let bufferFormat = AVAudioFormat(settings: outputFormatSettings)

    let outputBuffer = AVAudioPCMBuffer(pcmFormat: bufferFormat!, frameCapacity: AVAudioFrameCount(buff.count))

    for i in 0..<buff.count {
        outputBuffer?.floatChannelData!.pointee[i] = Float(( buff[i] ))
    }
    outputBuffer!.frameLength = AVAudioFrameCount( buff.count )

    do{
        try audioFile?.write(from: outputBuffer!)

    } catch let error as NSError {
        print("error:", error.localizedDescription)
    }
}

如果我运行这个应用程序，控制台将打印出创建的两个波形的 url(一个是生成的正弦波，另一个是记录的麦克风输入)。如果我在 daw 中检查这些，我会得到以下信息。您可以看到两个正弦波不同步。这让我相信采样率是不同的，但是打印到控制台的格式告诉我它们没有不同。

最初 inputNode 直接连接到 micSinkNode，但是我插入了一个 AVAudioMixerNode 以尝试在使用 AVAudioSinkNode 之前转换格式。

目标是能够使用任何使用自己的设置运行的 sampleRate 硬件，并将样本保存到应用程序首选的“ native 设置”。 (即应用程序将以 48kHz 的频率进行数字压缩。我希望能够使用 96k 硬件和不同的 channel 数)。

任何人都可以建议为什么这不起作用？

Answer 1

AVAudioSinkNode 不支持格式转换。

它必须处理硬件输入格式的数据。

引用:来自 WWDC 2019 的演讲，此处约 3 分 55 秒: https://developer.apple.com/videos/play/wwdc2019/510/

附言我目前正在努力在设备上设置一个非默认采样率(我假设硬件支持它，因为我从可用采样率列表中选择它)，但即使在这种情况下非默认采样率，接收器完全停止工作(没有崩溃，只是没有被调用)。