MP3 streaming on iOS - iphone

I want to use OpenAL to play music in an iOS game. The music files are stored in mp3 format and I want to stream them using a buffer queue. I load audio data into the buffers using AudioFileReadPacketData(). However playing the buffers only gives me noise. It works perfectly for caf files, but not for mp3s. Did I miss some vital step in decoding the file?
Code I use to open the sound file:
- (void) openFile:(NSString*)fileName {
NSBundle *bundle = [NSBundle mainBundle];
CFURLRef url = (CFURLRef)[[NSURL fileURLWithPath:[bundle pathForResource:fileName ofType:#"mp3"]] retain];
AudioFileOpenURL(url, kAudioFileReadPermission, 0, &audioFile);
AudioStreamBasicDescription theFormat;
UInt32 formatSize = sizeof(theFormat);
AudioFileGetProperty(audioFile, kAudioFilePropertyDataFormat, &formatSize, &theFormat);
freq = (ALsizei)theFormat.mSampleRate;
CFRelease(url);
}
Code I use to fill in buffers:
- (void) loadOneChunkIntoBuffer:(ALuint)buffer {
char data[STREAM_BUFFER_SIZE];
UInt32 loadSize = STREAM_BUFFER_SIZE;
AudioStreamPacketDescription packetDesc[STREAM_PACKETS];
UInt32 numPackets = STREAM_PACKETS;
AudioFileReadPacketData(audioFile, NO, &loadSize, packetDesc, packetsLoaded, &numPackets, data);
alBufferData(buffer, AL_FORMAT_STEREO16, data, loadSize, freq);
packetsLoaded += numPackets;
}

Because you're reading bytes of MP3 data and treating them as PCM data.
You almost certainly want AudioFileReadPacketData(). EDIT: Except that still gives you MP3 data; it just gives it in packets and (possibly) parses packet headers.
If you don't require OpenAL, AVAudioPlayer is probably the better way to go (according to the Multimedia Programming Guide, there's also Audio Queue services if you want more control).
If you really need to use OpenAL, according to TN2199 you'll need to convert it to PCM in the native byte order. See oalTouch/Classes/MyOpenALSupport.c for an example of using Extended Audio File Services to do this. Note that TN2199 says the format "must ... not use hardware decompression" — according to the Multimedia Programming Guide, software decoding is supported for everything except HE-AAC since OS 3.0. Also note that software MP3 decoding can use a significant amount of CPU time.
Alternatively, explicitly convert the audio using AudioConverter or (possibly) AudioUnit with kAudioUnitSubType_AUConverter. If you do this, it might be worthwhile decompressing everything once and keeping it in memory to minimize overhead.

Related

(iPhone) Live FFT from iPod

Okay guys, I've read many things about the FFT stuff, but it seems to be a bit more complicated than building a tableView.
I am searching for a way to analyze the playing audio (from iPod Library) in three ranges (low, mid, high). I think FFT is doing the job, but I'm not sure if I could filter (Lowpass, Bandpass and Highpass) the playing audio and analyze the peaks as well.
So if anyone knows what is the best (by best I mean, fastest (CPU) way to do so, please help me. There will be no front-end, so I won't draw the FFT in a Window (I guess the drawing does eat a lot of the cpu).
Then I have no idea how I could analyze the audio. All the FFT Sample Codes I found are using the mic. I do not want to use the mic. I saw something getting the Audio File and exporting it to a uncompressed file, but I need a live-analysation.
I've had a look at aurioTouch2, but I don't get how I could change the input from the mic to the iPod Library.
I think, the part I'm searching for is here:
// Initialize our remote i/o unit
inputProc.inputProc = PerformThru;
inputProc.inputProcRefCon = self;
CFURLRef url = NULL;
try {
url = CFURLCreateWithFileSystemPath(kCFAllocatorDefault, CFStringRef([[NSBundle mainBundle] pathForResource:#"button_press" ofType:#"caf"]), kCFURLPOSIXPathStyle, false);
XThrowIfError(AudioServicesCreateSystemSoundID(url, &buttonPressSound), "couldn't create button tap alert sound");
CFRelease(url);
// Initialize and configure the audio session
XThrowIfError(AudioSessionInitialize(NULL, NULL, rioInterruptionListener, self), "couldn't initialize audio session");
UInt32 audioCategory = kAudioSessionCategory_PlayAndRecord;
XThrowIfError(AudioSessionSetProperty(kAudioSessionProperty_AudioCategory, sizeof(audioCategory), &audioCategory), "couldn't set audio category");
XThrowIfError(AudioSessionAddPropertyListener(kAudioSessionProperty_AudioRouteChange, propListener, self), "couldn't set property listener");
Float32 preferredBufferSize = .005;
XThrowIfError(AudioSessionSetProperty(kAudioSessionProperty_PreferredHardwareIOBufferDuration, sizeof(preferredBufferSize), &preferredBufferSize), "couldn't set i/o buffer duration");
UInt32 size = sizeof(hwSampleRate);
XThrowIfError(AudioSessionGetProperty(kAudioSessionProperty_CurrentHardwareSampleRate, &size, &hwSampleRate), "couldn't get hw sample rate");
XThrowIfError(AudioSessionSetActive(true), "couldn't set audio session active\n");
XThrowIfError(SetupRemoteIO(rioUnit, inputProc, thruFormat), "couldn't setup remote i/o unit");
unitHasBeenCreated = true;
drawFormat.SetAUCanonical(2, false);
drawFormat.mSampleRate = 44100;
(...)
But I'm quite new to all of these AudioUnits, so I can't understand where an input is loaded. Then, the code mentioned above uses AVAudioSession. A little birdie told me, this will be deprecated, so what is the alternative?
So, basically:
How can I get the currently playing audio in order to do an analyzation? Can I just use a MPMusicPlayerController and get the samples? Or do I have to build a entire AudioUnit which plays the Library?
What is the fastest way (CPU) to analyze lows, mids and highs? Filtering? FFT? Something else?
Will I get in trouble with the Copyrights of bought music? Because I tried to convert the playing file to PCA Samples and sometimes I have this error:
VTM_AViPodReader[7666:307] * Terminating app
due to uncaught exception 'NSInvalidArgumentException', reason:
'* -[AVAssetReader initWithAsset:error:] invalid parameter not
satisfying: asset != ((void *)0)'
What is the "new" way to do an FFT if the whole AVAudioSession stuff won't work in the future?
You can't get the currently playing audio (security sandbox prevents this) on iOS, unless your app is the one playing the audio using certain select APIs (Audio Queue, RemoteIO, etc.)
3 bandpass filters (made with IIR biquads) will be faster than an FFT. But even a full FFT will use a very small percentage of CPU time.
An app can't convert or play protected music from the iTunes library in a form where samples can be captured.
The FFT is in the Accelerate framework, not in the audio session.

how can i play pcm with AVPlayer

I'm using AVPlayer to play local .mp3 file and audio stream from server.
And i want to play local .pcm file too.
NSArray *paths=NSSearchPathForDirectoriesInDomains(NSDocumentDirectory
, NSUserDomainMask
, YES);
NSString * voiceFile = [[paths objectAtIndex:0] stringByAppendingPathComponent:#"OutPut.pcm"];
But it didn't work. i got unknown error.
It seems AudioQueue can play .pcm correctly.
But is there a sample way can let AVPlayer direct play .pcm just like .mp3?
Neither .pcm as a file extension or PCM data specifies a readable format. The player cannot recognize an arbitrary data stream. It is certainly capable of reading file formats which contain PCM data, but this PCM representation is missing several things typical audio file formats represent:
Sample Rate
Sample Size
Sample Format
Channel Count
and so on.
You should instead save that PCM data in an audio file format the player supports (e.g. a WAV file).
If you prefer to simply stream PCM audio information and you know the stream format, you can approach that problem using an AudioQueue.

AudioToolbox/OpenAL ExtAudioFile to play compressed audio

I'm currently using OpenAL to play game music. It works fine, except that it doesn't work with anything except for raw WAV files. This means that I end up with a ~9mb soundtrack.
I'm new to OpenAL, and I'm using code directly from Apple's example (https://developer.apple.com/library/ios/#samplecode/MusicCube/Listings/Classes_MyOpenALSupport_h.html%23//apple_ref/doc/uid/DTS40008978-Classes_MyOpenALSupport_h-DontLinkElementID_9) to get the buffer data.
Question: Is there any way to modify this function so it reads compressed audio and decodes it on the fly?
I'm not so worried about the audio file format, just as long as it can be played and is compressed (like mp3, aac, caf). The only reason I want to do this (obviously) is to reduce file size.
Edit: It seems that the problem is not so much in OpenAL as the method I'm using to get the buffer. The function at https://developer.apple.com/library/ios/#samplecode/MusicCube/Listings/Classes_MyOpenALSupport_h.html%23//apple_ref/doc/uid/DTS40008978-Classes_MyOpenALSupport_h-DontLinkElementID_9 uses AudioFileOpenURL and AudioFileReadBytes. Is there any way to get the framework to decode the audio for me using ExtAudioFileOpenURL and ExtAudioFileRead?
I have tried the code here: https://devforums.apple.com/message/10678#10678, but I don't know what to make of it. The function I use to get the buffer is at https://developer.apple.com/library/ios/#samplecode/MusicCube/Listings/Classes_MyOpenALSupport_h.html%23//apple_ref/doc/uid/DTS40008978-Classes_MyOpenALSupport_h-DontLinkElementID_9, and I haven't really modified it, so that's what I need to build on.
I've started a bounty because I really need this, hopefully someone can point me in the right direction.
You'll need to use audio services to load other formats. Bear in mind that OpenAL ONLY supports uncompressed PCM data, so any data you load needs to be uncompressed during load.
Here's some code that will load any format supported by iOS: https://github.com/kstenerud/ObjectAL-for-iPhone/blob/master/ObjectAL/ObjectAL/Support/OALAudioFile.m
If you want to stream compressed soundtrack-type audio, use AVAudioPlayer since it plays compressed audio straight from disk.
You don't need any third party library to open archived files. With a little help from AudioToolbox/AudioToolbox.h framework you can open and read the data of a .caf file which is a very good choice by the way (better than mp3 or ogg) in terms of performance (minimal CPU impact during decompression). So ,when the data gets to OpenAL it is already PCM, ready to fill the buffers. Here is some sample code on how you can achieve this:
-(void) prepareFiles:(NSString *) filePath{
// get the full path of the file
NSString* fileName = [[NSBundle mainBundle] pathForResource:filePath ofType:#"caf"];
// open the file using the custom created methods (see below)
AudioFileID fileID = [self openAudioFile:fileName];
preparedAudioFileSize = [self audioFileSize:fileID];
if (preparedAudioFile){
free(preparedAudioFile);
preparedAudioFile = nil;
}
else{
;
}
preparedAudioFile = malloc(preparedAudioFileSize);
//read the data from the file into soundOutData var
AudioFileReadBytes(fileID, false, 0, &preparedAudioFileSize, preparedAudioFile);
//close the file
AudioFileClose(fileID);
}
-(AudioFileID)openAudioFile:(NSString*)filePath
{
AudioFileID fileID;
NSURL * url = [NSURL fileURLWithPath:filePath];
OSStatus result = AudioFileOpenURL((CFURLRef)url, kAudioFileReadPermission, 0, &fileID);
if (result != noErr) {
NSLog(#"fail to open: %#",filePath);
}
else {
;
}
return fileID;
}
-(UInt32)audioFileSize:(AudioFileID)fileDescriptor
{
UInt64 outDataSize = 0;
UInt32 thePropSize = sizeof(UInt64);
OSStatus result = AudioFileGetProperty(fileDescriptor, kAudioFilePropertyAudioDataByteCount, &thePropSize, &outDataSize);
if(result != 0) NSLog(#"cannot find file size");
return (UInt32)outDataSize;
}
based on Karl's reply above, I made a minimal single c++ function which opens a file and gives you back a buffer of pcm audio ( suitable for OpenAL ) and all the info you need to create an OpenAL sound ( format, samplerate, buffersize etc ).
the two files you need are here:
https://gist.github.com/ofTheo/5171369
hope it helps!
theo
Try if this works: http://kcat.strangesoft.net/openal-tutorial.html
You might try to use a third party library to load a mp3-ogg into a char* buffer, and then give this buffer to openAL. That would solve the file size problem.
For ogg, you should find the libraries on their website
For mp3, I honestly don't know where to find a lightweight library which could do that. But that should exist.

Capturing and manipulating microphone audio with AVCaptureSession?

While there are plenty of tutorials for how to use AVCaptureSession to grab camera data, I can find no information (even on apple's dev network itself) on how to properly handle microphone data.
I have implemented AVCaptureAudioDataOutputSampleBufferDelegate, and I'm getting calls to my delegate, but I have no idea how the contents of the CMSampleBufferRef I get are formatted. Are the contents of the buffer one discrete sample? What are its properties? Where can these properties be set?
Video properties can be set using [AVCaptureVideoDataOutput setVideoSettings:], but there is no corresponding call for AVCaptureAudioDataOutput (no setAudioSettings or anything similar).
They are formatted as LPCM! You can verify this by getting the AudioStreamBasicDescription like so:
CMFormatDescriptionRef formatDescription = CMSampleBufferGetFormatDescription(sampleBuffer);
const AudioStreamBasicDescription *streamDescription = CMAudioFormatDescriptionGetStreamBasicDescription(formatDescription);
and then checking the stream description’s mFormatId.

AVAudioPlayer - Metering - Want to build a waveform (graph)

I need to build a visual graph that represents voice levels (dB) in a recorded file. I tried to do it this way:
NSError *error = nil;
AVAudioPlayer *meterPlayer = [[AVAudioPlayer alloc]initWithContentsOfURL:[NSURL fileURLWithPath:self.recording.fileName] error:&error];
if (error) {
_lcl_logger(lcl_cEditRecording, lcl_vError, #"Cannot initialize AVAudioPlayer with file %# due to: %# (%#)", self.recording.fileName, error, error.userInfo);
} else {
[meterPlayer prepareToPlay];
meterPlayer.meteringEnabled = YES;
for (NSTimeInterval i = 0; i <= meterPlayer.duration; ++i) {
meterPlayer.currentTime = i;
[meterPlayer updateMeters];
float averagePower = [meterPlayer averagePowerForChannel:0];
_lcl_logger(lcl_cEditRecording, lcl_vTrace, #"Second: %f, Level: %f dB", i, averagePower);
}
}
[meterPlayer release];
It would be cool if it worked out however it didn't. I always get -160 dB. Any other ideas on how to implement that?
UPD: Here is what I got finally:
alt text http://img22.imageshack.us/img22/5778/waveform.png
I just want to help the others who have come into this same question and used a lot of time to search. To save your time, I put out my answer. I dislike somebody here who treat this as kind of secret...
After search around the articles about extaudioservice, audio queue and avfoundation.
I realised that i should use AVFoundation, reason is simple, it is the latest bundle and it is Objective C but not so cpp style.
So the steps to do it is not complicated:
Create AVAsset from the audio file
Create avassetreader from the avasset
Create avassettrack from avasset
Create avassetreadertrackoutput from avassettrack
Add the avassetreadertrackoutput to the previous avassetreader to start reading out the audio data
From the avassettrackoutput you can copyNextSampleBuffer one by one (it is a loop to read all data out).
Each copyNextSampleBuffer gives you a CMSampleBufferRef which can be used to get AudioBufferList by CMSampleBufferGetAudioBufferListWithRetainedBlockBuffer. AudioBufferList is array of AudioBuffer. AudioBuffer is the a bunch of audio data which is stored in its mData part.
You can implement the above in extAudioService as well. But i think the above avfoundation approach is easier.
So next question, what to do with the mData? Note that when you get the avassetreadertrackoutput, you can specify its output format, so we specify the output is lpcm.
Then the mData you finally get is actually a float format amplitude value.
Easy right? Though i used a lot of time to organise this from piece here and there.
Two useful resource for share:
Read this article to know basic terms and conceptions: https://www.mikeash.com/pyblog/friday-qa-2012-10-12-obtaining-and-interpreting-audio-data.html
Sample code: https://github.com/iluvcapra/JHWaveform
You can copy most of the above mentioned code from this sample directly and used for your own purpose.
I haven't used it myself, but Apple's avTouch iPhone sample has bar graphs powered by AVAudioPlayer, and you can easily check to see how they do it.
I don't think you can use AVAudioPlayer based on your constraints. Even if you could get it to "start" without actually playing the sound file, it would only help you build a graph as fast as the audio file would stream. What you're talking about is doing static analysis of the sound, which will require a much different approach. You'll need to read in the file yourself and parse it manually. I don't think there's a quick solution using anything in the SDK.
Ok guys, seems I'm going to answer my own question again: http://www.supermegaultragroovy.com/blog/2009/10/06/drawing-waveforms/ No a lot of concretics, but at least you will know what Apple docs to read.