/// <summary>
/// Saving the decoded Packets to our active Buffer, if the Buffer is full queue it into the OutputQueue
/// and wait until another buffer gets freed up
/// </summary>
private void AudioPacketDecoded(object sender, PacketReceivedEventArgs args)
{
foreach (var p in args.PacketDescriptions)
{
currentByteCount += p.DataByteSize;
AudioStreamPacketDescription pd = p;
int left = bufferSize - currentBuffer.CurrentOffset;
if (left < pd.DataByteSize)
{
EnqueueBuffer();
WaitForBuffer();
}
AudioQueue.FillAudioData(currentBuffer.Buffer, currentBuffer.CurrentOffset, args.InputData, (int)pd.StartOffset, pd.DataByteSize);
// Set new offset for this packet
pd.StartOffset = currentBuffer.CurrentOffset;
// Add the packet to our Buffer
currentBuffer.PacketDescriptions.Add(pd);
// Add the Size so that we know how much is in the buffer
currentBuffer.CurrentOffset += pd.DataByteSize;
}
if ((fileStream != null && currentByteCount == fileStream.DataByteCount) || lastPacket)
{
EnqueueBuffer();
}
}
/// <summary>
/// Saving the decoded Packets to our active Buffer, if the Buffer is full queue it into the OutputQueue
/// and wait until another buffer gets freed up
/// </summary>
private void AudioPacketDecoded(object sender, PacketReceivedEventArgs args)
{
if (BitRate == 0)
{
BitRate = ((AudioFileStream)sender).BitRate;
}
foreach (var p in args.PacketDescriptions)
{
_currentByteCount += p.DataByteSize;
AudioStreamPacketDescription packetDescription = p;
int left = BufferSize - _currentBuffer.CurrentOffset;
if (left < packetDescription.DataByteSize)
{
EnqueueBuffer();
WaitForBuffer();
}
AudioQueue.FillAudioData(_currentBuffer.Buffer, _currentBuffer.CurrentOffset, args.InputData, (int)packetDescription.StartOffset, packetDescription.DataByteSize);
// Set new offset for this packet
packetDescription.StartOffset = _currentBuffer.CurrentOffset;
// Add the packet to our Buffer
_currentBuffer.PacketDescriptions.Add(packetDescription);
// Add the Size so that we know how much is in the buffer
_currentBuffer.CurrentOffset += packetDescription.DataByteSize;
}
if ((_audioFileStream != null && _currentByteCount == _audioFileStream.DataByteCount) || _lastPacket)
{
_lastPacket = false;
Console.WriteLine($"{nameof(AudioPacketDecoded)} This is the lastpacket");
EnqueueBuffer();
}
}
public void FillBuffer(AudioBufferList data, uint numberFrames, AudioStreamPacketDescription[] packetDescs)
{
uint numPackets = numberFrames;
int err =AudioConverterFillComplexBuffer(
_audioConverter,
complexInputDataProc,
GCHandle.ToIntPtr(_handle),
ref numPackets,
data,
packetDescs);
if(err != 0 || numPackets == 0) {
throw new InvalidOperationException(String.Format("Error code:{0}", err));
}
}
/// <summary>
/// Saving the decoded Packets to our active Buffer, if the Buffer is full queue it into the OutputQueue
/// and wait until another buffer gets freed up
/// </summary>
void AudioPacketDecoded(object sender, PacketReceivedEventArgs args)
{
foreach (var p in args.PacketDescriptions)
{
currentByteCount += p.DataByteSize;
AudioStreamPacketDescription pd = p;
int left = bufferSize - currentBuffer.CurrentOffset;
if (left < pd.DataByteSize)
{
EnqueueBuffer();
WaitForBuffer();
}
AudioQueue.FillAudioData(currentBuffer.Buffer, currentBuffer.CurrentOffset, args.InputData, (int)pd.StartOffset, pd.DataByteSize);
#if true
// Set new offset for this packet
pd.StartOffset = currentBuffer.CurrentOffset;
// Add the packet to our Buffer
currentBuffer.PacketDescriptions.Add(pd);
// Add the Size so that we know how much is in the buffer
currentBuffer.CurrentOffset += pd.DataByteSize;
#else
// Fill out the packet description
pdesc [packetsFilled] = pd;
pdesc [packetsFilled].StartOffset = bytesFilled;
bytesFilled += packetSize;
packetsFilled++;
var t = OutputQueue.CurrentTime;
Console.WriteLine("Time: {0}", t);
// If we filled out all of our packet descriptions, enqueue the buffer
if (pdesc.Length == packetsFilled)
{
EnqueueBuffer();
WaitForBuffer();
}
#endif
}
if ((fileStream != null && currentByteCount == fileStream.DataByteCount) || lastPacket)
{
EnqueueBuffer();
}
}
static int complexInputDataProc(
IntPtr inAudioConverrter,
ref uint ioNumberDataPackets,
AudioBufferList ioData,
ref AudioStreamPacketDescription[] outDataPacketDescription, //AudioStreamPacketDescription**
IntPtr inUserData
)
{
// getting audiounit instance
var handler = GCHandle.FromIntPtr(inUserData);
var inst = (_AudioConverter)handler.Target;
// evoke event handler with an argument
if (inst.EncoderCallback != null)
{
var args = new _AudioConverterEventArgs(
ioNumberDataPackets,
ioData,
outDataPacketDescription);
inst.EncoderCallback(inst, args);
}
return 0; // noerror
}
public static bool Convert(string input, string output, AudioFormatType targetFormat, AudioFileType containerType, Microsoft.Xna.Framework.Content.Pipeline.Audio.ConversionQuality quality)
{
CFUrl source = CFUrl.FromFile(input);
CFUrl dest = CFUrl.FromFile(output);
var dstFormat = new AudioStreamBasicDescription();
var sourceFile = AudioFile.Open(source, AudioFilePermission.Read);
AudioFormatType outputFormat = targetFormat;
// get the source data format
var srcFormat = (AudioStreamBasicDescription)sourceFile.DataFormat;
var outputSampleRate = 0;
switch (quality)
{
case Microsoft.Xna.Framework.Content.Pipeline.Audio.ConversionQuality.Low:
outputSampleRate = (int)Math.Max(8000, srcFormat.SampleRate / 2);
break;
default:
outputSampleRate = (int)Math.Max(8000, srcFormat.SampleRate);
break;
}
dstFormat.SampleRate = (outputSampleRate == 0 ? srcFormat.SampleRate : outputSampleRate); // set sample rate
if (outputFormat == AudioFormatType.LinearPCM)
{
// if the output format is PC create a 16-bit int PCM file format description as an example
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = srcFormat.ChannelsPerFrame;
dstFormat.BitsPerChannel = 16;
dstFormat.BytesPerPacket = dstFormat.BytesPerFrame = 2 * dstFormat.ChannelsPerFrame;
dstFormat.FramesPerPacket = 1;
dstFormat.FormatFlags = AudioFormatFlags.LinearPCMIsPacked | AudioFormatFlags.LinearPCMIsSignedInteger;
}
else
{
// compressed format - need to set at least format, sample rate and channel fields for kAudioFormatProperty_FormatInfo
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = (outputFormat == AudioFormatType.iLBC ? 1 : srcFormat.ChannelsPerFrame); // for iLBC num channels must be 1
// use AudioFormat API to fill out the rest of the description
var fie = AudioStreamBasicDescription.GetFormatInfo(ref dstFormat);
if (fie != AudioFormatError.None)
{
return(false);
}
}
var converter = AudioConverter.Create(srcFormat, dstFormat);
converter.InputData += HandleInputData;
// if the source has a cookie, get it and set it on the Audio Converter
ReadCookie(sourceFile, converter);
// get the actual formats back from the Audio Converter
srcFormat = converter.CurrentInputStreamDescription;
dstFormat = converter.CurrentOutputStreamDescription;
// if encoding to AAC set the bitrate to 192k which is a nice value for this demo
// kAudioConverterEncodeBitRate is a UInt32 value containing the number of bits per second to aim for when encoding data
if (dstFormat.Format == AudioFormatType.MPEG4AAC)
{
uint outputBitRate = 192000; // 192k
// ignore errors as setting may be invalid depending on format specifics such as samplerate
try {
converter.EncodeBitRate = outputBitRate;
} catch {
}
// get it back and print it out
outputBitRate = converter.EncodeBitRate;
}
// create the destination file
var destinationFile = AudioFile.Create(dest, containerType, dstFormat, AudioFileFlags.EraseFlags);
// set up source buffers and data proc info struct
afio = new AudioFileIO(32768);
afio.SourceFile = sourceFile;
afio.SrcFormat = srcFormat;
if (srcFormat.BytesPerPacket == 0)
{
// if the source format is VBR, we need to get the maximum packet size
// use kAudioFilePropertyPacketSizeUpperBound which returns the theoretical maximum packet size
// in the file (without actually scanning the whole file to find the largest packet,
// as may happen with kAudioFilePropertyMaximumPacketSize)
afio.SrcSizePerPacket = sourceFile.PacketSizeUpperBound;
// how many packets can we read for our buffer size?
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
// allocate memory for the PacketDescription structures describing the layout of each packet
afio.PacketDescriptions = new AudioStreamPacketDescription [afio.NumPacketsPerRead];
}
else
{
// CBR source format
afio.SrcSizePerPacket = srcFormat.BytesPerPacket;
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
// allocate memory for the PacketDescription structures describing the layout of each packet
afio.PacketDescriptions = new AudioStreamPacketDescription [afio.NumPacketsPerRead];
}
// set up output buffers
int outputSizePerPacket = dstFormat.BytesPerPacket; // this will be non-zero if the format is CBR
const int theOutputBufSize = 32768;
var outputBuffer = Marshal.AllocHGlobal(theOutputBufSize);
AudioStreamPacketDescription[] outputPacketDescriptions = null;
if (outputSizePerPacket == 0)
{
// if the destination format is VBR, we need to get max size per packet from the converter
outputSizePerPacket = (int)converter.MaximumOutputPacketSize;
}
// allocate memory for the PacketDescription structures describing the layout of each packet
outputPacketDescriptions = new AudioStreamPacketDescription [theOutputBufSize / outputSizePerPacket];
int numOutputPackets = theOutputBufSize / outputSizePerPacket;
// if the destination format has a cookie, get it and set it on the output file
WriteCookie(converter, destinationFile);
// write destination channel layout
if (srcFormat.ChannelsPerFrame > 2)
{
WriteDestinationChannelLayout(converter, sourceFile, destinationFile);
}
long totalOutputFrames = 0; // used for debugging
long outputFilePos = 0;
AudioBuffers fillBufList = new AudioBuffers(1);
bool error = false;
// loop to convert data
while (true)
{
// set up output buffer list
fillBufList [0] = new AudioBuffer()
{
NumberChannels = dstFormat.ChannelsPerFrame,
DataByteSize = theOutputBufSize,
Data = outputBuffer
};
// convert data
int ioOutputDataPackets = numOutputPackets;
var fe = converter.FillComplexBuffer(ref ioOutputDataPackets, fillBufList, outputPacketDescriptions);
// if interrupted in the process of the conversion call, we must handle the error appropriately
if (fe != AudioConverterError.None)
{
error = true;
break;
}
if (ioOutputDataPackets == 0)
{
// this is the EOF conditon
break;
}
// write to output file
var inNumBytes = fillBufList [0].DataByteSize;
var we = destinationFile.WritePackets(false, inNumBytes, outputPacketDescriptions, outputFilePos, ref ioOutputDataPackets, outputBuffer);
if (we != 0)
{
error = true;
break;
}
// advance output file packet position
outputFilePos += ioOutputDataPackets;
if (dstFormat.FramesPerPacket != 0)
{
// the format has constant frames per packet
totalOutputFrames += (ioOutputDataPackets * dstFormat.FramesPerPacket);
}
else
{
// variable frames per packet require doing this for each packet (adding up the number of sample frames of data in each packet)
for (var i = 0; i < ioOutputDataPackets; ++i)
{
totalOutputFrames += outputPacketDescriptions [i].VariableFramesInPacket;
}
}
}
Marshal.FreeHGlobal(outputBuffer);
if (!error)
{
// write out any of the leading and trailing frames for compressed formats only
if (dstFormat.BitsPerChannel == 0)
{
// our output frame count should jive with
WritePacketTableInfo(converter, destinationFile);
}
// write the cookie again - sometimes codecs will update cookies at the end of a conversion
WriteCookie(converter, destinationFile);
}
converter.Dispose();
destinationFile.Dispose();
sourceFile.Dispose();
return(true);
}
unsafe static void RenderAudio (CFUrl sourceUrl, CFUrl destinationUrl)
{
AudioStreamBasicDescription dataFormat;
AudioQueueBuffer *buffer = null;
long currentPacket = 0;
int packetsToRead = 0;
AudioStreamPacketDescription [] packetDescs = null;
bool flushed = false;
bool done = false;
int bufferSize;
using (var audioFile = AudioFile.Open (sourceUrl, AudioFilePermission.Read, (AudioFileType) 0)) {
dataFormat = audioFile.StreamBasicDescription;
using (var queue = new OutputAudioQueue (dataFormat, CFRunLoop.Current, CFRunLoop.CFRunLoopCommonModes)) {
queue.OutputCompleted += (sender, e) =>
{
HandleOutput (audioFile, queue, buffer, ref packetsToRead, ref currentPacket, ref done, ref flushed, ref packetDescs);
};
// we need to calculate how many packets we read at a time and how big a buffer we need
// we base this on the size of the packets in the file and an approximate duration for each buffer
bool isVBR = dataFormat.BytesPerPacket == 0 || dataFormat.FramesPerPacket == 0;
// first check to see what the max size of a packet is - if it is bigger
// than our allocation default size, that needs to become larger
// adjust buffer size to represent about a second of audio based on this format
CalculateBytesForTime (dataFormat, audioFile.MaximumPacketSize, 1.0, out bufferSize, out packetsToRead);
if (isVBR) {
packetDescs = new AudioStreamPacketDescription [packetsToRead];
} else {
packetDescs = null; // we don't provide packet descriptions for constant bit rate formats (like linear PCM)
}
if (audioFile.MagicCookie.Length != 0)
queue.MagicCookie = audioFile.MagicCookie;
// allocate the input read buffer
queue.AllocateBuffer (bufferSize, out buffer);
// prepare the capture format
var captureFormat = AudioStreamBasicDescription.CreateLinearPCM (dataFormat.SampleRate, (uint) dataFormat.ChannelsPerFrame, 32);
captureFormat.BytesPerFrame = captureFormat.BytesPerPacket = dataFormat.ChannelsPerFrame * 4;
queue.SetOfflineRenderFormat (captureFormat, audioFile.ChannelLayout);
// prepare the target format
var dstFormat = AudioStreamBasicDescription.CreateLinearPCM (dataFormat.SampleRate, (uint) dataFormat.ChannelsPerFrame);
using (var captureFile = ExtAudioFile.CreateWithUrl (destinationUrl, AudioFileType.CAF, dstFormat, AudioFileFlags.EraseFlags)) {
captureFile.ClientDataFormat = captureFormat;
int captureBufferSize = bufferSize / 2;
AudioBuffers captureABL = new AudioBuffers (1);
AudioQueueBuffer *captureBuffer;
queue.AllocateBuffer (captureBufferSize, out captureBuffer);
captureABL[0] = new AudioBuffer () {
Data = captureBuffer->AudioData,
NumberChannels = captureFormat.ChannelsPerFrame
};
queue.Start ();
double ts = 0;
queue.RenderOffline (ts, captureBuffer, 0);
HandleOutput (audioFile, queue, buffer, ref packetsToRead, ref currentPacket, ref done, ref flushed, ref packetDescs);
while (true) {
int reqFrames = captureBufferSize / captureFormat.BytesPerFrame;
queue.RenderOffline (ts, captureBuffer, reqFrames);
captureABL.SetData (0, captureBuffer->AudioData, (int) captureBuffer->AudioDataByteSize);
var writeFrames = captureABL[0].DataByteSize / captureFormat.BytesPerFrame;
// Console.WriteLine ("ts: {0} AudioQueueOfflineRender: req {1} frames / {2} bytes, got {3} frames / {4} bytes",
// ts, reqFrames, captureBufferSize, writeFrames, captureABL.Buffers [0].DataByteSize);
captureFile.WriteAsync ((uint) writeFrames, captureABL);
if (flushed)
break;
ts += writeFrames;
}
CFRunLoop.Current.RunInMode (CFRunLoop.CFDefaultRunLoopMode, 1, false);
}
}
}
}
unsafe static void HandleOutput (AudioFile audioFile, AudioQueue queue, AudioQueueBuffer *audioQueueBuffer, ref int packetsToRead, ref long currentPacket, ref bool done, ref bool flushed, ref AudioStreamPacketDescription [] packetDescriptions)
{
int bytes;
int packets;
if (done)
return;
packets = packetsToRead;
bytes = (int) audioQueueBuffer->AudioDataBytesCapacity;
packetDescriptions = audioFile.ReadPacketData (false, currentPacket, ref packets, audioQueueBuffer->AudioData, ref bytes);
if (packets > 0) {
audioQueueBuffer->AudioDataByteSize = (uint) bytes;
queue.EnqueueBuffer (audioQueueBuffer, packetDescriptions);
currentPacket += packets;
} else {
if (!flushed) {
queue.Flush ();
flushed = true;
}
queue.Stop (false);
done = true;
}
}
// Input data proc callback
AudioConverterError EncoderDataProc (ref int numberDataPackets, AudioBuffers data, ref AudioStreamPacketDescription[] dataPacketDescription)
{
// figure out how much to read
int maxPackets = afio.SrcBufferSize / afio.SrcSizePerPacket;
if (numberDataPackets > maxPackets)
numberDataPackets = maxPackets;
// read from the file
int outNumBytes;
var res = afio.SourceFile.ReadPackets (false, out outNumBytes, afio.PacketDescriptions, afio.SrcFilePos, ref numberDataPackets, afio.SrcBuffer);
if (res != 0) {
throw new ApplicationException (res.ToString ());
}
// advance input file packet position
afio.SrcFilePos += numberDataPackets;
// put the data pointer into the buffer list
data.SetData (0, afio.SrcBuffer, outNumBytes);
// don't forget the packet descriptions if required
if (dataPacketDescription != null) {
if (afio.PacketDescriptions != null) {
dataPacketDescription = afio.PacketDescriptions;
} else {
dataPacketDescription = null;
}
}
return AudioConverterError.None;
}
bool DoConvertFile (CFUrl sourceURL, NSUrl destinationURL, AudioFormatType outputFormat, double outputSampleRate)
{
AudioStreamBasicDescription dstFormat = new AudioStreamBasicDescription ();
// in this sample we should never be on the main thread here
Debug.Assert (!NSThread.IsMain);
// transition thread state to State::Running before continuing
AppDelegate.ThreadStateSetRunning ();
Debug.WriteLine ("DoConvertFile");
// get the source file
var sourceFile = AudioFile.Open (sourceURL, AudioFilePermission.Read);
// get the source data format
var srcFormat = (AudioStreamBasicDescription)sourceFile.DataFormat;
// setup the output file format
dstFormat.SampleRate = (outputSampleRate == 0 ? srcFormat.SampleRate : outputSampleRate); // set sample rate
if (outputFormat == AudioFormatType.LinearPCM) {
// if the output format is PC create a 16-bit int PCM file format description as an example
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = srcFormat.ChannelsPerFrame;
dstFormat.BitsPerChannel = 16;
dstFormat.BytesPerPacket = dstFormat.BytesPerFrame = 2 * dstFormat.ChannelsPerFrame;
dstFormat.FramesPerPacket = 1;
dstFormat.FormatFlags = AudioFormatFlags.LinearPCMIsPacked | AudioFormatFlags.LinearPCMIsSignedInteger;
} else {
// compressed format - need to set at least format, sample rate and channel fields for kAudioFormatProperty_FormatInfo
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = (outputFormat == AudioFormatType.iLBC ? 1 : srcFormat.ChannelsPerFrame); // for iLBC num channels must be 1
// use AudioFormat API to fill out the rest of the description
var fie = AudioStreamBasicDescription.GetFormatInfo (ref dstFormat);
if (fie != AudioFormatError.None) {
Debug.Print ("Cannot create destination format {0:x}", fie);
AppDelegate.ThreadStateSetDone ();
return false;
}
}
// create the AudioConverter
AudioConverterError ce;
var converter = AudioConverter.Create (srcFormat, dstFormat, out ce);
Debug.Assert (ce == AudioConverterError.None);
converter.InputData += EncoderDataProc;
// if the source has a cookie, get it and set it on the Audio Converter
ReadCookie (sourceFile, converter);
// get the actual formats back from the Audio Converter
srcFormat = converter.CurrentInputStreamDescription;
dstFormat = converter.CurrentOutputStreamDescription;
// if encoding to AAC set the bitrate to 192k which is a nice value for this demo
// kAudioConverterEncodeBitRate is a UInt32 value containing the number of bits per second to aim for when encoding data
if (dstFormat.Format == AudioFormatType.MPEG4AAC) {
uint outputBitRate = 192000; // 192k
// ignore errors as setting may be invalid depending on format specifics such as samplerate
try {
converter.EncodeBitRate = outputBitRate;
} catch {
}
// get it back and print it out
outputBitRate = converter.EncodeBitRate;
Debug.Print ("AAC Encode Bitrate: {0}", outputBitRate);
}
// can the Audio Converter resume conversion after an interruption?
// this property may be queried at any time after construction of the Audio Converter after setting its output format
// there's no clear reason to prefer construction time, interruption time, or potential resumption time but we prefer
// construction time since it means less code to execute during or after interruption time
bool canResumeFromInterruption;
try {
canResumeFromInterruption = converter.CanResumeFromInterruption;
Debug.Print ("Audio Converter {0} continue after interruption!", canResumeFromInterruption ? "CAN" : "CANNOT");
} catch (Exception e) {
// if the property is unimplemented (kAudioConverterErr_PropertyNotSupported, or paramErr returned in the case of PCM),
// then the codec being used is not a hardware codec so we're not concerned about codec state
// we are always going to be able to resume conversion after an interruption
canResumeFromInterruption = false;
Debug.Print ("CanResumeFromInterruption: {0}", e.Message);
}
// create the destination file
var destinationFile = AudioFile.Create (destinationURL, AudioFileType.CAF, dstFormat, AudioFileFlags.EraseFlags);
// set up source buffers and data proc info struct
afio = new AudioFileIO (32768);
afio.SourceFile = sourceFile;
afio.SrcFormat = srcFormat;
if (srcFormat.BytesPerPacket == 0) {
// if the source format is VBR, we need to get the maximum packet size
// use kAudioFilePropertyPacketSizeUpperBound which returns the theoretical maximum packet size
// in the file (without actually scanning the whole file to find the largest packet,
// as may happen with kAudioFilePropertyMaximumPacketSize)
afio.SrcSizePerPacket = sourceFile.PacketSizeUpperBound;
// how many packets can we read for our buffer size?
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
// allocate memory for the PacketDescription structures describing the layout of each packet
afio.PacketDescriptions = new AudioStreamPacketDescription [afio.NumPacketsPerRead];
} else {
// CBR source format
afio.SrcSizePerPacket = srcFormat.BytesPerPacket;
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
}
// set up output buffers
int outputSizePerPacket = dstFormat.BytesPerPacket; // this will be non-zero if the format is CBR
const int theOutputBufSize = 32768;
var outputBuffer = Marshal.AllocHGlobal (theOutputBufSize);
AudioStreamPacketDescription[] outputPacketDescriptions = null;
if (outputSizePerPacket == 0) {
// if the destination format is VBR, we need to get max size per packet from the converter
outputSizePerPacket = (int)converter.MaximumOutputPacketSize;
// allocate memory for the PacketDescription structures describing the layout of each packet
outputPacketDescriptions = new AudioStreamPacketDescription [theOutputBufSize / outputSizePerPacket];
}
int numOutputPackets = theOutputBufSize / outputSizePerPacket;
// if the destination format has a cookie, get it and set it on the output file
WriteCookie (converter, destinationFile);
// write destination channel layout
if (srcFormat.ChannelsPerFrame > 2) {
WriteDestinationChannelLayout (converter, sourceFile, destinationFile);
}
long totalOutputFrames = 0; // used for debugging
long outputFilePos = 0;
AudioBuffers fillBufList = new AudioBuffers (1);
bool error = false;
// loop to convert data
Debug.WriteLine ("Converting...");
while (true) {
// set up output buffer list
fillBufList [0] = new AudioBuffer () {
NumberChannels = dstFormat.ChannelsPerFrame,
DataByteSize = theOutputBufSize,
Data = outputBuffer
};
// this will block if we're interrupted
var wasInterrupted = AppDelegate.ThreadStatePausedCheck();
if (wasInterrupted && !canResumeFromInterruption) {
// this is our interruption termination condition
// an interruption has occured but the Audio Converter cannot continue
Debug.WriteLine ("Cannot resume from interruption");
error = true;
break;
}
// convert data
int ioOutputDataPackets = numOutputPackets;
var fe = converter.FillComplexBuffer (ref ioOutputDataPackets, fillBufList, outputPacketDescriptions);
// if interrupted in the process of the conversion call, we must handle the error appropriately
if (fe != AudioConverterError.None) {
Debug.Print ("FillComplexBuffer: {0}", fe);
error = true;
break;
}
if (ioOutputDataPackets == 0) {
// this is the EOF conditon
break;
}
// write to output file
var inNumBytes = fillBufList [0].DataByteSize;
var we = destinationFile.WritePackets (false, inNumBytes, outputPacketDescriptions, outputFilePos, ref ioOutputDataPackets, outputBuffer);
if (we != 0) {
Debug.Print ("WritePackets: {0}", we);
error = true;
break;
}
// advance output file packet position
outputFilePos += ioOutputDataPackets;
if (dstFormat.FramesPerPacket != 0) {
// the format has constant frames per packet
totalOutputFrames += (ioOutputDataPackets * dstFormat.FramesPerPacket);
} else {
// variable frames per packet require doing this for each packet (adding up the number of sample frames of data in each packet)
for (var i = 0; i < ioOutputDataPackets; ++i)
totalOutputFrames += outputPacketDescriptions [i].VariableFramesInPacket;
}
}
Marshal.FreeHGlobal (outputBuffer);
if (!error) {
// write out any of the leading and trailing frames for compressed formats only
if (dstFormat.BitsPerChannel == 0) {
// our output frame count should jive with
Debug.Print ("Total number of output frames counted: {0}", totalOutputFrames);
WritePacketTableInfo (converter, destinationFile);
}
// write the cookie again - sometimes codecs will update cookies at the end of a conversion
WriteCookie (converter, destinationFile);
}
converter.Dispose ();
destinationFile.Dispose ();
sourceFile.Dispose ();
// transition thread state to State.Done before continuing
AppDelegate.ThreadStateSetDone ();
return !error;
}
bool DoConvertFile(CFUrl sourceURL, NSUrl destinationURL, AudioFormatType outputFormat, double outputSampleRate)
{
AudioStreamBasicDescription dstFormat = new AudioStreamBasicDescription();
// in this sample we should never be on the main thread here
Debug.Assert(!NSThread.IsMain);
// transition thread state to State::Running before continuing
AppDelegate.ThreadStateSetRunning();
Debug.WriteLine("DoConvertFile");
// get the source file
var sourceFile = AudioFile.Open(sourceURL, AudioFilePermission.Read);
// get the source data format
var srcFormat = (AudioStreamBasicDescription)sourceFile.DataFormat;
// setup the output file format
dstFormat.SampleRate = (outputSampleRate == 0 ? srcFormat.SampleRate : outputSampleRate); // set sample rate
if (outputFormat == AudioFormatType.LinearPCM)
{
// if the output format is PC create a 16-bit int PCM file format description as an example
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = srcFormat.ChannelsPerFrame;
dstFormat.BitsPerChannel = 16;
dstFormat.BytesPerPacket = dstFormat.BytesPerFrame = 2 * dstFormat.ChannelsPerFrame;
dstFormat.FramesPerPacket = 1;
dstFormat.FormatFlags = AudioFormatFlags.LinearPCMIsPacked | AudioFormatFlags.LinearPCMIsSignedInteger;
}
else
{
// compressed format - need to set at least format, sample rate and channel fields for kAudioFormatProperty_FormatInfo
dstFormat.Format = outputFormat;
dstFormat.ChannelsPerFrame = (outputFormat == AudioFormatType.iLBC ? 1 : srcFormat.ChannelsPerFrame); // for iLBC num channels must be 1
// use AudioFormat API to fill out the rest of the description
var fie = AudioStreamBasicDescription.GetFormatInfo(ref dstFormat);
if (fie != AudioFormatError.None)
{
Debug.Print("Cannot create destination format {0:x}", fie);
AppDelegate.ThreadStateSetDone();
return(false);
}
}
// create the AudioConverter
AudioConverterError ce;
var converter = AudioConverter.Create(srcFormat, dstFormat, out ce);
Debug.Assert(ce == AudioConverterError.None);
converter.InputData += EncoderDataProc;
// if the source has a cookie, get it and set it on the Audio Converter
ReadCookie(sourceFile, converter);
// get the actual formats back from the Audio Converter
srcFormat = converter.CurrentInputStreamDescription;
dstFormat = converter.CurrentOutputStreamDescription;
// if encoding to AAC set the bitrate to 192k which is a nice value for this demo
// kAudioConverterEncodeBitRate is a UInt32 value containing the number of bits per second to aim for when encoding data
if (dstFormat.Format == AudioFormatType.MPEG4AAC)
{
uint outputBitRate = 192000; // 192k
// ignore errors as setting may be invalid depending on format specifics such as samplerate
try {
converter.EncodeBitRate = outputBitRate;
} catch {
}
// get it back and print it out
outputBitRate = converter.EncodeBitRate;
Debug.Print("AAC Encode Bitrate: {0}", outputBitRate);
}
// can the Audio Converter resume conversion after an interruption?
// this property may be queried at any time after construction of the Audio Converter after setting its output format
// there's no clear reason to prefer construction time, interruption time, or potential resumption time but we prefer
// construction time since it means less code to execute during or after interruption time
bool canResumeFromInterruption;
try {
canResumeFromInterruption = converter.CanResumeFromInterruption;
Debug.Print("Audio Converter {0} continue after interruption!", canResumeFromInterruption ? "CAN" : "CANNOT");
} catch (Exception e) {
// if the property is unimplemented (kAudioConverterErr_PropertyNotSupported, or paramErr returned in the case of PCM),
// then the codec being used is not a hardware codec so we're not concerned about codec state
// we are always going to be able to resume conversion after an interruption
canResumeFromInterruption = false;
Debug.Print("CanResumeFromInterruption: {0}", e.Message);
}
// create the destination file
var destinationFile = AudioFile.Create(destinationURL, AudioFileType.CAF, dstFormat, AudioFileFlags.EraseFlags);
// set up source buffers and data proc info struct
afio = new AudioFileIO(32768);
afio.SourceFile = sourceFile;
afio.SrcFormat = srcFormat;
if (srcFormat.BytesPerPacket == 0)
{
// if the source format is VBR, we need to get the maximum packet size
// use kAudioFilePropertyPacketSizeUpperBound which returns the theoretical maximum packet size
// in the file (without actually scanning the whole file to find the largest packet,
// as may happen with kAudioFilePropertyMaximumPacketSize)
afio.SrcSizePerPacket = sourceFile.PacketSizeUpperBound;
// how many packets can we read for our buffer size?
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
// allocate memory for the PacketDescription structures describing the layout of each packet
afio.PacketDescriptions = new AudioStreamPacketDescription [afio.NumPacketsPerRead];
}
else
{
// CBR source format
afio.SrcSizePerPacket = srcFormat.BytesPerPacket;
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
}
// set up output buffers
int outputSizePerPacket = dstFormat.BytesPerPacket; // this will be non-zero if the format is CBR
const int theOutputBufSize = 32768;
var outputBuffer = Marshal.AllocHGlobal(theOutputBufSize);
AudioStreamPacketDescription[] outputPacketDescriptions = null;
if (outputSizePerPacket == 0)
{
// if the destination format is VBR, we need to get max size per packet from the converter
outputSizePerPacket = (int)converter.MaximumOutputPacketSize;
// allocate memory for the PacketDescription structures describing the layout of each packet
outputPacketDescriptions = new AudioStreamPacketDescription [theOutputBufSize / outputSizePerPacket];
}
int numOutputPackets = theOutputBufSize / outputSizePerPacket;
// if the destination format has a cookie, get it and set it on the output file
WriteCookie(converter, destinationFile);
// write destination channel layout
if (srcFormat.ChannelsPerFrame > 2)
{
WriteDestinationChannelLayout(converter, sourceFile, destinationFile);
}
long totalOutputFrames = 0; // used for debugging
long outputFilePos = 0;
AudioBuffers fillBufList = new AudioBuffers(1);
bool error = false;
// loop to convert data
Debug.WriteLine("Converting...");
while (true)
{
// set up output buffer list
fillBufList [0] = new AudioBuffer()
{
NumberChannels = dstFormat.ChannelsPerFrame,
DataByteSize = theOutputBufSize,
Data = outputBuffer
};
// this will block if we're interrupted
var wasInterrupted = AppDelegate.ThreadStatePausedCheck();
if (wasInterrupted && !canResumeFromInterruption)
{
// this is our interruption termination condition
// an interruption has occured but the Audio Converter cannot continue
Debug.WriteLine("Cannot resume from interruption");
error = true;
break;
}
// convert data
int ioOutputDataPackets = numOutputPackets;
var fe = converter.FillComplexBuffer(ref ioOutputDataPackets, fillBufList, outputPacketDescriptions);
// if interrupted in the process of the conversion call, we must handle the error appropriately
if (fe != AudioConverterError.None)
{
Debug.Print("FillComplexBuffer: {0}", fe);
error = true;
break;
}
if (ioOutputDataPackets == 0)
{
// this is the EOF conditon
break;
}
// write to output file
var inNumBytes = fillBufList [0].DataByteSize;
var we = destinationFile.WritePackets(false, inNumBytes, outputPacketDescriptions, outputFilePos, ref ioOutputDataPackets, outputBuffer);
if (we != 0)
{
Debug.Print("WritePackets: {0}", we);
error = true;
break;
}
// advance output file packet position
outputFilePos += ioOutputDataPackets;
if (dstFormat.FramesPerPacket != 0)
{
// the format has constant frames per packet
totalOutputFrames += (ioOutputDataPackets * dstFormat.FramesPerPacket);
}
else
{
// variable frames per packet require doing this for each packet (adding up the number of sample frames of data in each packet)
for (var i = 0; i < ioOutputDataPackets; ++i)
{
totalOutputFrames += outputPacketDescriptions [i].VariableFramesInPacket;
}
}
}
Marshal.FreeHGlobal(outputBuffer);
if (!error)
{
// write out any of the leading and trailing frames for compressed formats only
if (dstFormat.BitsPerChannel == 0)
{
// our output frame count should jive with
Debug.Print("Total number of output frames counted: {0}", totalOutputFrames);
WritePacketTableInfo(converter, destinationFile);
}
// write the cookie again - sometimes codecs will update cookies at the end of a conversion
WriteCookie(converter, destinationFile);
}
converter.Dispose();
destinationFile.Dispose();
sourceFile.Dispose();
// transition thread state to State.Done before continuing
AppDelegate.ThreadStateSetDone();
return(!error);
}
unsafe static void RenderAudio(CFUrl sourceUrl, CFUrl destinationUrl)
{
AudioStreamBasicDescription dataFormat;
AudioQueueBuffer * buffer = null;
long currentPacket = 0;
int packetsToRead = 0;
AudioStreamPacketDescription[] packetDescs = null;
bool flushed = false;
bool done = false;
int bufferSize;
using (var audioFile = AudioFile.Open(sourceUrl, AudioFilePermission.Read, (AudioFileType)0)) {
dataFormat = audioFile.StreamBasicDescription;
using (var queue = new OutputAudioQueue(dataFormat, CFRunLoop.Current, CFRunLoop.ModeCommon)) {
queue.BufferCompleted += (sender, e) => {
HandleOutput(audioFile, queue, buffer, ref packetsToRead, ref currentPacket, ref done, ref flushed, ref packetDescs);
};
// we need to calculate how many packets we read at a time and how big a buffer we need
// we base this on the size of the packets in the file and an approximate duration for each buffer
bool isVBR = dataFormat.BytesPerPacket == 0 || dataFormat.FramesPerPacket == 0;
// first check to see what the max size of a packet is - if it is bigger
// than our allocation default size, that needs to become larger
// adjust buffer size to represent about a second of audio based on this format
CalculateBytesForTime(dataFormat, audioFile.MaximumPacketSize, 1.0, out bufferSize, out packetsToRead);
if (isVBR)
{
packetDescs = new AudioStreamPacketDescription [packetsToRead];
}
else
{
packetDescs = null; // we don't provide packet descriptions for constant bit rate formats (like linear PCM)
}
if (audioFile.MagicCookie.Length != 0)
{
queue.MagicCookie = audioFile.MagicCookie;
}
// allocate the input read buffer
queue.AllocateBuffer(bufferSize, out buffer);
// prepare the capture format
var captureFormat = AudioStreamBasicDescription.CreateLinearPCM(dataFormat.SampleRate, (uint)dataFormat.ChannelsPerFrame, 32);
captureFormat.BytesPerFrame = captureFormat.BytesPerPacket = dataFormat.ChannelsPerFrame * 4;
queue.SetOfflineRenderFormat(captureFormat, audioFile.ChannelLayout);
// prepare the target format
var dstFormat = AudioStreamBasicDescription.CreateLinearPCM(dataFormat.SampleRate, (uint)dataFormat.ChannelsPerFrame);
using (var captureFile = ExtAudioFile.CreateWithUrl(destinationUrl, AudioFileType.CAF, dstFormat, AudioFileFlags.EraseFlags)) {
captureFile.ClientDataFormat = captureFormat;
int captureBufferSize = bufferSize / 2;
AudioBuffers captureABL = new AudioBuffers(1);
AudioQueueBuffer *captureBuffer;
queue.AllocateBuffer(captureBufferSize, out captureBuffer);
captureABL [0] = new AudioBuffer()
{
Data = captureBuffer->AudioData,
NumberChannels = captureFormat.ChannelsPerFrame
};
queue.Start();
double ts = 0;
queue.RenderOffline(ts, captureBuffer, 0);
HandleOutput(audioFile, queue, buffer, ref packetsToRead, ref currentPacket, ref done, ref flushed, ref packetDescs);
while (true)
{
int reqFrames = captureBufferSize / captureFormat.BytesPerFrame;
queue.RenderOffline(ts, captureBuffer, reqFrames);
captureABL.SetData(0, captureBuffer->AudioData, (int)captureBuffer->AudioDataByteSize);
var writeFrames = captureABL [0].DataByteSize / captureFormat.BytesPerFrame;
// Console.WriteLine ("ts: {0} AudioQueueOfflineRender: req {1} frames / {2} bytes, got {3} frames / {4} bytes",
// ts, reqFrames, captureBufferSize, writeFrames, captureABL.Buffers [0].DataByteSize);
captureFile.WriteAsync((uint)writeFrames, captureABL);
if (flushed)
{
break;
}
ts += writeFrames;
}
CFRunLoop.Current.RunInMode(CFRunLoop.ModeDefault, 1, false);
}
}
}
}
void Convert(string sourceFilePath, string destinationFilePath, AudioFormatType outputFormatType, int?sampleRate = null)
{
var destinationUrl = NSUrl.FromFilename(destinationFilePath);
var sourceUrl = NSUrl.FromFilename(sourceFilePath);
// get the source file
var name = Path.GetFileNameWithoutExtension(destinationFilePath);
using var sourceFile = AudioFile.Open(sourceUrl, AudioFilePermission.Read);
var srcFormat = (AudioStreamBasicDescription)sourceFile.DataFormat;
var dstFormat = new AudioStreamBasicDescription();
// setup the output file format
dstFormat.SampleRate = sampleRate ?? srcFormat.SampleRate;
if (outputFormatType == AudioFormatType.LinearPCM)
{
// if the output format is PCM create a 16 - bit int PCM file format
dstFormat.Format = AudioFormatType.LinearPCM;
dstFormat.ChannelsPerFrame = srcFormat.ChannelsPerFrame;
dstFormat.BitsPerChannel = 16;
dstFormat.BytesPerPacket = dstFormat.BytesPerFrame = 2 * dstFormat.ChannelsPerFrame;
dstFormat.FramesPerPacket = 1;
dstFormat.FormatFlags = AudioFormatFlags.LinearPCMIsPacked | AudioFormatFlags.LinearPCMIsSignedInteger;
}
else
{
// compressed format - need to set at least format, sample rate and channel fields for kAudioFormatProperty_FormatInfo
dstFormat.Format = outputFormatType;
dstFormat.ChannelsPerFrame = srcFormat.ChannelsPerFrame; // for iLBC num channels must be 1
// use AudioFormat API to fill out the rest of the description
var afe = AudioStreamBasicDescription.GetFormatInfo(ref dstFormat);
Assert.AreEqual(AudioFormatError.None, afe, $"GetFormatInfo: {name}");
}
// create the AudioConverter
using var converter = AudioConverter.Create(srcFormat, dstFormat, out var ce);
Assert.AreEqual(AudioConverterError.None, ce, $"AudioConverterCreate: {name}");
// set up source buffers and data proc info struct
var afio = new AudioFileIO(32 * 1024); // 32Kb
converter.InputData += (ref int numberDataPackets, AudioBuffers data, ref AudioStreamPacketDescription [] dataPacketDescription) => {
return(EncoderDataProc(afio, ref numberDataPackets, data, ref dataPacketDescription));
};
// Some audio formats have a magic cookie associated with them which is required to decompress audio data
// When converting audio data you must check to see if the format of the data has a magic cookie
// If the audio data format has a magic cookie associated with it, you must add this information to anAudio Converter
// using AudioConverterSetProperty and kAudioConverterDecompressionMagicCookie to appropriately decompress the data
// http://developer.apple.com/mac/library/qa/qa2001/qa1318.html
var cookie = sourceFile.MagicCookie;
// if there is an error here, then the format doesn't have a cookie - this is perfectly fine as some formats do not
if (cookie?.Length > 0)
{
converter.DecompressionMagicCookie = cookie;
}
// get the actual formats back from the Audio Converter
srcFormat = converter.CurrentInputStreamDescription;
dstFormat = converter.CurrentOutputStreamDescription;
// create the destination file
using var destinationFile = AudioFile.Create(destinationUrl, AudioFileType.CAF, dstFormat, AudioFileFlags.EraseFlags);
// set up source buffers and data proc info struct
afio.SourceFile = sourceFile;
afio.SrcFormat = srcFormat;
if (srcFormat.BytesPerPacket == 0)
{
// if the source format is VBR, we need to get the maximum packet size
// use kAudioFilePropertyPacketSizeUpperBound which returns the theoretical maximum packet size
// in the file (without actually scanning the whole file to find the largest packet,
// as may happen with kAudioFilePropertyMaximumPacketSize)
afio.SrcSizePerPacket = sourceFile.PacketSizeUpperBound;
// how many packets can we read for our buffer size?
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
// allocate memory for the PacketDescription structures describing the layout of each packet
afio.PacketDescriptions = new AudioStreamPacketDescription [afio.NumPacketsPerRead];
}
else
{
// CBR source format
afio.SrcSizePerPacket = srcFormat.BytesPerPacket;
afio.NumPacketsPerRead = afio.SrcBufferSize / afio.SrcSizePerPacket;
}
// set up output buffers
int outputSizePerPacket = dstFormat.BytesPerPacket; // this will be non-zero if the format is CBR
const int theOutputBufSize = 32 * 1024; // 32Kb
var outputBuffer = Marshal.AllocHGlobal(theOutputBufSize);
AudioStreamPacketDescription [] outputPacketDescriptions = null;
if (outputSizePerPacket == 0)
{
// if the destination format is VBR, we need to get max size per packet from the converter
outputSizePerPacket = (int)converter.MaximumOutputPacketSize;
// allocate memory for the PacketDescription structures describing the layout of each packet
outputPacketDescriptions = new AudioStreamPacketDescription [theOutputBufSize / outputSizePerPacket];
}
int numOutputPackets = theOutputBufSize / outputSizePerPacket;
// if the destination format has a cookie, get it and set it on the output file
WriteCookie(converter, destinationFile);
long totalOutputFrames = 0; // used for debugging
long outputFilePos = 0;
AudioBuffers fillBufList = new AudioBuffers(1);
// loop to convert data
while (true)
{
// set up output buffer list
fillBufList [0] = new AudioBuffer()
{
NumberChannels = dstFormat.ChannelsPerFrame,
DataByteSize = theOutputBufSize,
Data = outputBuffer
};
// convert data
int ioOutputDataPackets = numOutputPackets;
var fe = converter.FillComplexBuffer(ref ioOutputDataPackets, fillBufList, outputPacketDescriptions);
// if interrupted in the process of the conversion call, we must handle the error appropriately
Assert.AreEqual(AudioConverterError.None, fe, $"FillComplexBuffer: {name}");
if (ioOutputDataPackets == 0)
{
// this is the EOF conditon
break;
}
// write to output file
var inNumBytes = fillBufList [0].DataByteSize;
var we = destinationFile.WritePackets(false, inNumBytes, outputPacketDescriptions, outputFilePos, ref ioOutputDataPackets, outputBuffer);
Assert.AreEqual(AudioFileError.Success, we, $"WritePackets: {name}");
// advance output file packet position
outputFilePos += ioOutputDataPackets;
// the format has constant frames per packet
totalOutputFrames += (ioOutputDataPackets * dstFormat.FramesPerPacket);
}
Marshal.FreeHGlobal(outputBuffer);
// write out any of the leading and trailing frames for compressed formats only
if (dstFormat.BitsPerChannel == 0)
{
WritePacketTableInfo(converter, destinationFile);
}
// write the cookie again - sometimes codecs will update cookies at the end of a conversion
WriteCookie(converter, destinationFile);
}
// Input data proc callback
AudioConverterError EncoderDataProc(ref int numberDataPackets, AudioBuffers data, ref AudioStreamPacketDescription[] dataPacketDescription)
{
// figure out how much to read
int maxPackets = afio.SrcBufferSize / afio.SrcSizePerPacket;
if (numberDataPackets > maxPackets)
numberDataPackets = maxPackets;
// read from the file
int outNumBytes = 16384;
// modified for iOS7 (ReadPackets depricated)
afio.PacketDescriptions = afio.SourceFile.ReadPacketData(false, afio.SrcFilePos, ref numberDataPackets, afio.SrcBuffer, ref outNumBytes);
if (afio.PacketDescriptions.Length == 0 && numberDataPackets > 0)
throw new ApplicationException(afio.PacketDescriptions.ToString());
// advance input file packet position
afio.SrcFilePos += numberDataPackets;
// put the data pointer into the buffer list
data.SetData(0, afio.SrcBuffer, outNumBytes);
// don't forget the packet descriptions if required
if (dataPacketDescription != null)
dataPacketDescription = afio.PacketDescriptions;
return AudioConverterError.None;
}
public unsafe void ConvertData(AudioBuffers audioBufferList, uint frames, float[][] buffers, out AudioStreamPacketDescription packetDescriptions)
{
var totalArrays = buffers.Length;
var arraySize = buffers [0].Length;
var pSize = Marshal.SizeOf(typeof(IntPtr));
IntPtr pDataBuffers;
IntPtr pPointerBuffers;
IntPtrHelper.ConvertToIntPtr <float> (buffers, out pDataBuffers, out pPointerBuffers);
_ConvertData((IntPtr)audioBufferList, frames, pPointerBuffers, out packetDescriptions);
for (int i = 0; i < totalArrays; i++)
{
var pArray = Marshal.ReadIntPtr(IntPtr.Add(pPointerBuffers, i * pSize));
fixed(float *arrAddr = &buffers[i][0])
{
IntPtrHelper.MemoryCopy((IntPtr)arrAddr, pArray, (nuint)arraySize);
}
}
Marshal.FreeHGlobal(pDataBuffers);
Marshal.FreeHGlobal(pPointerBuffers);
}
static extern int AudioConverterFillComplexBuffer(
IntPtr inAudioConverter,
AudioConverterComplexInputDataProc inInputDataProc,
IntPtr inInputDataProcUserData,
ref uint ioOutputDataPacketSize,
AudioBufferList outOutputData,
AudioStreamPacketDescription[] outPacketDescription);
