Slow buffering when streaming multiple remote videos with AVPlayer and AVMutableComposition - ios

[Edit: I was able to figure out a workaround for this, see below.]
I'm trying to stream multiple remote MP4 clips from S3, playing them in a sequence as one continuous video (to enable scrubbing within and between clips) with no stuttering, without explicitly downloading them to the device first. However, I find that the clips buffer very slowly (even on a fast network connection) and have been unable to find an adequate way to address that.
I've been trying to use AVPlayer for this, since AVPlayer with AVMutableComposition plays the supplied video tracks as one continuous track (unlike AVQueuePlayer, which I gather plays each video separately and thus doesn't support continuous scrubbing between the clips).
When I stick one of the assets directly into an AVPlayerItem and play that (with no AVMutableComposition), it buffers fast. But using AVMutableComposition, the video starts stuttering very badly on the second clip (my test case has 6 clips, each around 6 seconds), while the audio keeps going. After it plays through once, it plays perfectly smoothly if I rewind to the beginning, so I assume the problem lies in the buffering.
My current attempt to fix this problem feels convoluted, given that this seems like a rather basic use-case for AVPlayer - I do hope there's a simpler solution to all this that works properly. Somehow I doubt that the buffering player I use below is really necessary, but I'm running out of ideas.
Here's the main code that sets up the AVMutableComposition:
// Build an AVAsset for each of the source URIs
- (void)prepareAssetsForSources:(NSArray *)sources
{
NSMutableArray *assets = [[NSMutableArray alloc] init]; // the assets to be used in the AVMutableComposition
NSMutableArray *offsets = [[NSMutableArray alloc] init]; // for tracking buffering progress
CMTime currentOffset = kCMTimeZero;
for (NSDictionary* source in sources) {
bool isNetwork = [RCTConvert BOOL:[source objectForKey:#"isNetwork"]];
bool isAsset = [RCTConvert BOOL:[source objectForKey:#"isAsset"]];
NSString *uri = [source objectForKey:#"uri"];
NSString *type = [source objectForKey:#"type"];
NSURL *url = isNetwork ?
[NSURL URLWithString:uri] :
[[NSURL alloc] initFileURLWithPath:[[NSBundle mainBundle] pathForResource:uri ofType:type]];
AVURLAsset *asset = [AVURLAsset URLAssetWithURL:url options:nil];
currentOffset = CMTimeAdd(currentOffset, asset.duration);
[assets addObject:asset];
[offsets addObject:[NSNumber numberWithFloat:CMTimeGetSeconds(currentOffset)]];
}
_clipAssets = assets;
_clipEndOffsets = offsets;
}
// Called with _clipAssets
- (AVPlayerItem*)playerItemForAssets:(NSMutableArray *)assets
{
AVMutableComposition* composition = [AVMutableComposition composition];
for (AVAsset* asset in assets) {
CMTimeRange editRange = CMTimeRangeMake(CMTimeMake(0, 600), asset.duration);
NSError *editError;
[composition insertTimeRange:editRange
ofAsset:asset
atTime:composition.duration
error:&editError];
}
AVPlayerItem* playerItem = [AVPlayerItem playerItemWithAsset:composition];
return playerItem; // this is used to initialize the main player
}
My initial thought was: Since it buffers fast with a vanilla AVPlayerItem, why not maintain a separate buffering player that's loaded with each asset in turn (with no AVMutableComposition) to buffer the assets for the main player?
- (void)startBufferingClips
{
_bufferingPlayerItem = [AVPlayerItem playerItemWithAsset:_clipAssets[0]
automaticallyLoadedAssetKeys:#[#"tracks"]];
_bufferingPlayer = [AVPlayer playerWithPlayerItem:_bufferingPlayerItem];
_currentlyBufferingIndex = 0;
}
// called every 250 msecs via an addPeriodicTimeObserverForInterval on the main player
- (void)updateBufferingProgress
{
// If the playable (loaded) range is within 100 milliseconds of the clip
// currently being buffered, load the next clip into the buffering player.
float playableDuration = [[self calculateBufferedDuration] floatValue];
CMTime totalDurationTime = [self playerItemDuration :_bufferingPlayer];
Float64 totalDurationSeconds = CMTimeGetSeconds(totalDurationTime);
bool bufferingComplete = totalDurationSeconds - playableDuration < 0.1;
float bufferedSeconds = [self bufferedSeconds :playableDuration];
float playerTimeSeconds = CMTimeGetSeconds([_player currentTime]);
__block NSUInteger playingClipIndex = 0;
// find the index of _player's currently playing clip
[_clipEndOffsets enumerateObjectsUsingBlock:^(id offset, NSUInteger idx, BOOL *stop) {
if (playerTimeSeconds < [offset floatValue]) {
playingClipIndex = idx;
*stop = YES;
}
}];
// TODO: if bufferedSeconds - playerTimeSeconds <= 0, pause the main player
if (bufferingComplete && _currentlyBufferingIndex < [_clipAssets count] - 1) {
// We're done buffering this clip, load the buffering player with the next asset
_currentlyBufferingIndex += 1;
_bufferingPlayerItem = [AVPlayerItem playerItemWithAsset:_clipAssets[_currentlyBufferingIndex]
automaticallyLoadedAssetKeys:#[#"tracks"]];
_bufferingPlayer = [AVPlayer playerWithPlayerItem:_bufferingPlayerItem];
}
}
- (float)bufferedSeconds:(float)playableDuration {
__block float seconds = 0.0; // total duration of clips already buffered
if (_currentlyBufferingIndex > 0) {
[_clipEndOffsets enumerateObjectsUsingBlock:^(id offset, NSUInteger idx, BOOL *stop) {
if (idx + 1 >= _currentlyBufferingIndex) {
seconds = [offset floatValue];
*stop = YES;
}
}];
}
return seconds + playableDuration;
}
- (NSNumber *)calculateBufferedDuration {
AVPlayerItem *video = _bufferingPlayer.currentItem;
if (video.status == AVPlayerItemStatusReadyToPlay) {
__block float longestPlayableRangeSeconds;
[video.loadedTimeRanges enumerateObjectsUsingBlock:^(id obj, NSUInteger idx, BOOL *stop) {
CMTimeRange timeRange = [obj CMTimeRangeValue];
float seconds = CMTimeGetSeconds(CMTimeRangeGetEnd(timeRange));
if (seconds > 0.1) {
if (!longestPlayableRangeSeconds) {
longestPlayableRangeSeconds = seconds;
} else if (seconds > longestPlayableRangeSeconds) {
longestPlayableRangeSeconds = seconds;
}
}
}];
Float64 playableDuration = longestPlayableRangeSeconds;
if (playableDuration && playableDuration > 0) {
return [NSNumber numberWithFloat:longestPlayableRangeSeconds];
}
}
return [NSNumber numberWithInteger:0];
}
It initially seemed that this worked like a charm, but then I switched to another set of test clips and then the buffering was very slow again (the buffering player helped, but not enough). It seems like the loadedTimeRanges for the assets as loaded into the buffering player didn't match the loadedTimeRanges for the same assets inside the AVMutableComposition: Even after the loadedTimeRanges for each item loaded into the buffering player indicated that the whole asset had been buffered, the main player's video continued stuttering (while the audio played seamlessly to the end). Again, the playback was seamless after rewinding once the main player had run through all the clips once.
I hope the answer to this, whatever it is, will prove useful as a starting point for other iOS developers trying to implement this basic use-case. Thanks!
Edit: Since I posted this question, I made the following workaround for this. Hopefully this will save whoever runs into this some headache.
What I ended up doing was maintaining two buffering players (both AVPlayers) that started buffering the first two clips, moving on to the lowest-indexed unbuffered clip after their loadedTimeRanges indicated that buffering for their current clip was done. I made the logic pause/unpause playback based on the clips currently buffered, and the loadedTimeRanges of the buffering players, plus a small margin. This needed a few bookkeeping variables, but wasn't too complicated.
This is how the buffering players were initialized (I'm omitting the bookkeeping logic here):
- (void)startBufferingClips
{
_bufferingPlayerItemA = [AVPlayerItem playerItemWithAsset:_clipAssets[0]
automaticallyLoadedAssetKeys:#[#"tracks"]];
_bufferingPlayerA = [AVPlayer playerWithPlayerItem:_bufferingPlayerItemA];
_currentlyBufferingIndexA = [NSNumber numberWithInt:0];
if ([_clipAssets count] > 1) {
_bufferingPlayerItemB = [AVPlayerItem playerItemWithAsset:_clipAssets[1]
automaticallyLoadedAssetKeys:#[#"tracks"]];
_bufferingPlayerB = [AVPlayer playerWithPlayerItem:_bufferingPlayerItemB];
_currentlyBufferingIndexB = [NSNumber numberWithInt:1];
_nextIndexToBuffer = [NSNumber numberWithInt:2];
} else {
_nextIndexToBuffer = [NSNumber numberWithInt:1];
}
}
In addition, I needed to make sure that the video and audio tracks weren't being merged as they were added to AVMutableComposition, as this apparently interfered with the buffering (perhaps they didn't register as the same video/audio tracks as those the buffering players were loading, and thus didn't receive the new data). Here's the code where the AVMutableComposition is built from an array of NSAssets:
- (AVPlayerItem*)playerItemForAssets:(NSMutableArray *)assets
{
AVMutableComposition* composition = [AVMutableComposition composition];
AVMutableCompositionTrack *compVideoTrack = [composition addMutableTrackWithMediaType:AVMediaTypeVideo
preferredTrackID:kCMPersistentTrackID_Invalid];
AVMutableCompositionTrack *compAudioTrack = [composition addMutableTrackWithMediaType:AVMediaTypeAudio
preferredTrackID:kCMPersistentTrackID_Invalid];
CMTime timeOffset = kCMTimeZero;
for (AVAsset* asset in assets) {
CMTimeRange editRange = CMTimeRangeMake(CMTimeMake(0, 600), asset.duration);
NSError *editError;
NSArray *videoTracks = [asset tracksWithMediaType:AVMediaTypeVideo];
NSArray *audioTracks = [asset tracksWithMediaType:AVMediaTypeAudio];
if ([videoTracks count] > 0) {
AVAssetTrack *videoTrack = [videoTracks objectAtIndex:0];
[compVideoTrack insertTimeRange:editRange
ofTrack:videoTrack
atTime:timeOffset
error:&editError];
}
if ([audioTracks count] > 0) {
AVAssetTrack *audioTrack = [audioTracks objectAtIndex:0];
[compAudioTrack insertTimeRange:editRange
ofTrack:audioTrack
atTime:timeOffset
error:&editError];
}
if ([videoTracks count] > 0 || [audioTracks count] > 0) {
timeOffset = CMTimeAdd(timeOffset, asset.duration);
}
}
AVPlayerItem* playerItem = [AVPlayerItem playerItemWithAsset:composition];
return playerItem;
}
With this approach, buffering while using AVMutableComposition for the main player works nice and fast, at least in my setup.

Related

How to save the position of the audio file? objective -C

How can I save time when audio was stopped in session and continue playback from the stop point in next session?
My code:
- (void)initPlayer:(NSString*) audioFile fileExtension:(NSString*)fileExtension
{
NSURL *audioFileLocationURL = [[NSBundle mainBundle] URLForResource:audioFile withExtension:fileExtension];
NSError *error;
self.audioPlayer = [[AVAudioPlayer alloc] initWithContentsOfURL:audioFileLocationURL error:&error];
if ([audioFile isEqualToString:#"2"]) {
_index = 1;
}
else if ([audioFile isEqualToString:#"3"]) {
_index = 2;
}
[self song];
}
- (void)playAudio {
[self.audioPlayer play];
}
- (void)pauseAudio {
[self.audioPlayer pause];
}
- (BOOL)isPlaying {
return [self.audioPlayer isPlaying];
}
-(NSString*)timeFormat:(float)value{
float minutes = floor(lroundf(value)/60);
float seconds = lroundf(value) - (minutes * 60);
int roundedSeconds = lroundf(seconds);
int roundedMinutes = lroundf(minutes);
NSString *time = [[NSString alloc]
initWithFormat:#"%d:%02d",
roundedMinutes, roundedSeconds];
return time;
}
- (void)setCurrentAudioTime:(float)value {
[self.audioPlayer setCurrentTime:value];
}
- (NSTimeInterval)getCurrentAudioTime {
return [self.audioPlayer currentTime];
}
- (float)getAudioDuration {
return [self.audioPlayer duration];
}
You can use AVPlayer's currentTime property. It returns the playback time of the current AVPlayerItem.
To restore the playback time in the next session, you can pass the stored time to AVPlayer's seekToTime:
[self.player seekToTime:storedPlaybackTime];
https://developer.apple.com/library/mac/documentation/AVFoundation/Reference/AVPlayer_Class/index.html#//apple_ref/doc/uid/TP40009530-CH1-SW2
https://developer.apple.com/library/mac/documentation/AVFoundation/Reference/AVPlayer_Class/index.html#//apple_ref/occ/instm/AVPlayer/seekToTime%3a
To persist the CMTime returned by currentTime, you can use the AVFoundation convenience methods provided by NSValue.
To wrap CMTime in an NSValue, use valueWithCMTime:
[NSValue valueWithCMTime:player.currentTime];
To get an CMTime struct from the persisted value, use:
CMTime persistedTime = [storeValue CMTimeValue];
After you wrapped the CMTime struct in a NSValue instance, you can use keyed archiver & NSData to write the time to disk.
NSHipster has a good article about that topic:http://nshipster.com/nscoding/
The most easy way will be to keep a local db with the song name and when it is stopped add that data to the db. Then when the playback resumes later check the local db first if it has any entries in the past. If not continue from starting.
Also make sure that there is no entry made if the song finishes.
Hope this idea helps you...

precise timing with AVMutableComposition

I'm trying to use AVMutableComposition to play a sequence of sound files at precise times.
When the view loads, I create the composition with the intent of playing 4 sounds evenly spaced over 1 second. It shouldn't matter how long or short the sounds are, I just want to fire them at exactly 0, 0.25, 0.5 and 0.75 seconds:
AVMutableComposition *composition = [[AVMutableComposition alloc] init];
NSDictionary *options = #{AVURLAssetPreferPreciseDurationAndTimingKey : #YES};
for (NSInteger i = 0; i < 4; i++)
{
AVMutableCompositionTrack* track = [composition addMutableTrackWithMediaType:AVMediaTypeAudio preferredTrackID:kCMPersistentTrackID_Invalid];
NSURL *url = [[NSBundle mainBundle] URLForResource:[NSString stringWithFormat:#"sound_file_%i", i] withExtension:#"caf"];
AVURLAsset *asset = [AVURLAsset URLAssetWithURL:url options:options];
AVAssetTrack *assetTrack = [asset tracksWithMediaType:AVMediaTypeAudio].firstObject;
CMTimeRange timeRange = [assetTrack timeRange];
Float64 t = i * 0.25;
NSError *error;
BOOL success = [track insertTimeRange:timeRange ofTrack:assetTrack atTime:CMTimeMakeWithSeconds(t, 1) error:&error];
if (!success)
{
NSLog(#"unsuccesful creation of composition");
}
if (error)
{
NSLog(#"composition creation error: %#", error);
}
}
AVPlayerItem* playerItem = [AVPlayerItem playerItemWithAsset:composition];
self.avPlayer = [[AVPlayer alloc] initWithPlayerItem:playerItem];
The composition is created successfully with no errors. Later, when I want to play the sequence I do this:
[self.avPlayer seekToTime:CMTimeMakeWithSeconds(0, 1)];
[self.avPlayer play];
For some reason, the sounds are not evenly spaced at all - but play almost all at once. I tried the same thing spaced over 4 seconds, replacing the time calculation like this:
Float64 t = i * 1.0;
And this plays perfectly. Any time interval under 1 second seems to generate unexpected results. What am I missing? Are AVCompositions not supposed to be used for time intervals under 1 second? Or perhaps I'm misunderstanding the time intervals?
Your CMTimeMakeWithSeconds(t, 1) is in whole second 'slices' because your timescale is set to 1. No matter what fraction t is, the atTime: will always end up as 0. This is why it works when you increase it to 1 second (t=i*1).
You need to set the timescale to 4 to get your desired 0.25 second slices. Since the CMTime is now in .25 second slices, you won't need the i * 0.25 calculcation. Just use the i directly; atTime:CMTimeMake(i, 4)
If you might need to get more precise in the future, you should account for it now so you won't have to adjust your code later. Apple recommends using a timescale of 600 as it is a multiple of the common video framerates (24, 25, and 30 FPS) but it works fine for audio-only too. So for your situation, you would use 24 slices to get your .25 second value; Float64 t = i * 24; atTime:CMTimeMake(t, 600)
As for your issue of all 4 sounds playing almost all at once, be aware of this unanswered SO question where it only happens on the first play. Even with the changes above, you might still run into this issue.
Unless each track is exactly 0.25 seconds long this is your problem:
Float64 t = i * 0.25;
NSError *error;
BOOL success = [track insertTimeRange:timeRange ofTrack:assetTrack atTime:CMTimeMakeWithSeconds(t, 1) error:&error];
You need to be keeping track of the cumulative time range added so far, and inserting the next track at that time:
CMTime currentTime = kCMTimeZero;
for (NSInteger i = 0; i < 4; i++) {
/* Code to create track for insertion */
CMTimeRange trackTimeRange = [assetTrack timeRange];
BOOL success = [track insertTimeRange:trackTimeRange
ofTrack:assetTrack
atTime:currentTime
error:&error];
/* Error checking code */
//Update time range for insertion
currentTime = CMTimeAdd(currentTime,trackTimeRange.duration);
}
i changed a bit your code, sorry i had no time to test it.
AVMutableComposition *composition = [AVMutableComposition composition];
NSDictionary *options = #{AVURLAssetPreferPreciseDurationAndTimingKey : #YES};
CMTime totalDuration = kCMTimeZero;
for (NSInteger i = 0; i < 4; i++)
{
AVMutableCompositionTrack* track = [composition addMutableTrackWithMediaType:AVMediaTypeAudio preferredTrackID:kCMPersistentTrackID_Invalid];
NSURL *url = [NSURL fileURLWithPath:[[NSBundle mainBundle] pathForResource:[NSString stringWithFormat:#"Record_%i", i] ofType:#"caf"]];
AVURLAsset *asset = [[AVURLAsset alloc] initWithURL:url options:options];
AVAssetTrack *assetTrack = [[asset tracksWithMediaType:AVMediaTypeAudio] objectAtIndex:0];
CMTimeRange timeRange = [assetTrack timeRange];
NSError *error;
BOOL success = [track insertTimeRange:timeRange ofTrack:assetTrack atTime:CMTIME_COMPARE_INLINE(totalDuration, >, kCMTimeZero)? CMTimeAdd(totalDuration, CMTimeMake(1, 4)): totalDuration error:&error];
if (!success)
{
NSLog(#"unsuccesful creation of composition");
}
if (error)
{
NSLog(#"composition creation error: %#", error);
}
totalDuration = CMTimeAdd(CMTimeAdd(totalDuration,CMTimeMake(1, 4)), asset.duration);
}
AVPlayerItem* playerItem = [AVPlayerItem playerItemWithAsset:composition];
self.avPlayer = [[AVPlayer alloc] initWithPlayerItem:playerItem];
P.S. use kCMTimeZero instead of CMTimeMakeWithSeconds(0, 1).

HTTP live stream AVAsset

I am implementing a n HTTP live streaming player with OSX using avplayer.
I am able to stream it properly seek and get duration timing etc.
Now i want to take screen shots and process the frames from it using OpenCV.
I went for using AVASSetImageGenerator. But there is no audio and video tracks with the AVAsset which is associated with player.currentItem.
The tracks are appearing in player.currentItem.tracks.
So i am not able to sue AVAssetGenerator. Can anybody help to find out a solution to extract screenshots and individual frames in such a scenario?
Please find the code below how i am initiating an HTTP live stream
Thanks in advance.
NSURL* url = [NSURL URLWithString:#"http://devimages.apple.com/iphone/samples/bipbop/bipbopall.m3u8"];
playeritem = [AVPlayerItem playerItemWithURL:url];
[playeritem addObserver:self forKeyPath:#"status" options:0 context:AVSPPlayerStatusContext];
[self setPlayer:[AVPlayer playerWithPlayerItem:playeritem]];
[self addObserver:self forKeyPath:#"player.rate" options:NSKeyValueObservingOptionNew context:AVSPPlayerRateContext];
[self addObserver:self forKeyPath:#"player.currentItem.status" options:NSKeyValueObservingOptionNew context:AVSPPlayerItemStatusContext];
AVPlayerLayer *newPlayerLayer = [AVPlayerLayer playerLayerWithPlayer:[self player]];
[newPlayerLayer setFrame:[[[self playerView] layer] bounds]];
[newPlayerLayer setAutoresizingMask:kCALayerWidthSizable | kCALayerHeightSizable];
[newPlayerLayer setHidden:YES];
[[[self playerView] layer] addSublayer:newPlayerLayer];
[self setPlayerLayer:newPlayerLayer];
[self addObserver:self forKeyPath:#"playerLayer.readyForDisplay" options:NSKeyValueObservingOptionInitial | NSKeyValueObservingOptionNew context:AVSPPlayerLayerReadyForDisplay];
[self.player play];
Following is how i am checking whether video track is present with the Asset
case AVPlayerItemStatusReadyToPlay:
[self setTimeObserverToken:[[self player] addPeriodicTimeObserverForInterval:CMTimeMake(1, 10) queue:dispatch_get_main_queue() usingBlock:^(CMTime time) {
[[self timeSlider] setDoubleValue:CMTimeGetSeconds(time)];
NSLog(#"%f,%f,%f",[self currentTime],[self duration],[[self player] rate]);
AVPlayerItem *item = playeritem;
if(item.status == AVPlayerItemStatusReadyToPlay)
{
AVAsset *asset = (AVAsset *)item.asset;
long audiotracks = [[asset tracks] count];
long videotracks = [[asset availableMediaCharacteristicsWithMediaSelectionOptions]count];
NSLog(#"Track info Audio = %ld,Video=%ld",audiotracks,videotracks);
}
}]];
AVPlayerItem *item = self.player.currentItem;
if(item.status != AVPlayerItemStatusReadyToPlay)
return;
AVURLAsset *asset = (AVURLAsset *)item.asset;
long audiotracks = [[asset tracksWithMediaType:AVMediaTypeAudio]count];
long videotracks = [[asset tracksWithMediaType:AVMediaTypeVideo]count];
NSLog(#"Track info Audio = %ld,Video=%ld",audiotracks,videotracks);
This is an older question but in case someone needs help for that i have an answer
AVURLAsset *asset = /* Your Asset here! */;
AVAssetImageGenerator *generator = [[AVAssetImageGenerator alloc] initWithAsset:asset];
generator.requestedTimeToleranceAfter = kCMTimeZero;
generator.requestedTimeToleranceBefore = kCMTimeZero;
for (Float64 i = 0; i < CMTimeGetSeconds(asset.duration) * /* Put the FPS of the source video here */ ; i++){
#autoreleasepool {
CMTime time = CMTimeMake(i, /* Put the FPS of the source video here */);
NSError *err;
CMTime actualTime;
CGImageRef image = [generator copyCGImageAtTime:time actualTime:&actualTime error:&err];
// Do what you want with the image, for example save it as UIImage
UIImage *generatedImage = [[UIImage alloc] initWithCGImage:image];
CGImageRelease(image);
}
}
You can easily get the FPS of a Video by using this code:
float fps=0.00;
if (asset) {
AVAssetTrack * videoATrack = [asset tracksWithMediaType:AVMediaTypeVideo][0];
if(videoATrack)
{
fps = [videoATrack nominalFrameRate];
}
}
Hope that helps someone who is asking how to get all frames from a video or just some specific (with CMTime for example) frames. Please bear in mind, that saving all frames to an array can impact the memory hardly!

iOS: How to trim silence from start and end of .aif audio recording?

My app includes the ability for the user to record a brief message; I'd like to trim off any silence (or, to be more precise, any audio whose volume falls below a given threshold) from the beginning and end of the recording.
I'm recording the audio with an AVAudioRecorder, and saving it to an .aif file. I've seen some mention elsewhere of methods by which I could have it wait to start recording until the audio level reaches a threshold; that'd get me halfway there, but won't help with trimming silence off the end.
If there's a simple way to do this, I'll be eternally grateful!
Thanks.
This project takes audio from the microphone, triggers on loud noise and untriggers when quiet. It also trims and fades in/fades out around the ends.
https://github.com/fulldecent/FDSoundActivatedRecorder
Relevant code you are seeking:
- (NSString *)recordedFilePath
{
// Prepare output
NSString *trimmedAudioFileBaseName = [NSString stringWithFormat:#"recordingConverted%x.caf", arc4random()];
NSString *trimmedAudioFilePath = [NSTemporaryDirectory() stringByAppendingPathComponent:trimmedAudioFileBaseName];
NSFileManager *fileManager = [NSFileManager defaultManager];
if ([fileManager fileExistsAtPath:trimmedAudioFilePath]) {
NSError *error;
if ([fileManager removeItemAtPath:trimmedAudioFilePath error:&error] == NO) {
NSLog(#"removeItemAtPath %# error:%#", trimmedAudioFilePath, error);
}
}
NSLog(#"Saving to %#", trimmedAudioFilePath);
AVAsset *avAsset = [AVAsset assetWithURL:self.audioRecorder.url];
NSArray *tracks = [avAsset tracksWithMediaType:AVMediaTypeAudio];
AVAssetTrack *track = [tracks objectAtIndex:0];
AVAssetExportSession *exportSession = [AVAssetExportSession
exportSessionWithAsset:avAsset
presetName:AVAssetExportPresetAppleM4A];
// create trim time range
CMTime startTime = CMTimeMake(self.recordingBeginTime*SAVING_SAMPLES_PER_SECOND, SAVING_SAMPLES_PER_SECOND);
CMTimeRange exportTimeRange = CMTimeRangeFromTimeToTime(startTime, kCMTimePositiveInfinity);
// create fade in time range
CMTime startFadeInTime = startTime;
CMTime endFadeInTime = CMTimeMake(self.recordingBeginTime*SAVING_SAMPLES_PER_SECOND + RISE_TRIGGER_INTERVALS*INTERVAL_SECONDS*SAVING_SAMPLES_PER_SECOND, SAVING_SAMPLES_PER_SECOND);
CMTimeRange fadeInTimeRange = CMTimeRangeFromTimeToTime(startFadeInTime, endFadeInTime);
// setup audio mix
AVMutableAudioMix *exportAudioMix = [AVMutableAudioMix audioMix];
AVMutableAudioMixInputParameters *exportAudioMixInputParameters =
[AVMutableAudioMixInputParameters audioMixInputParametersWithTrack:track];
[exportAudioMixInputParameters setVolumeRampFromStartVolume:0.0 toEndVolume:1.0
timeRange:fadeInTimeRange];
exportAudioMix.inputParameters = [NSArray
arrayWithObject:exportAudioMixInputParameters];
// configure export session output with all our parameters
exportSession.outputURL = [NSURL fileURLWithPath:trimmedAudioFilePath];
exportSession.outputFileType = AVFileTypeAppleM4A;
exportSession.timeRange = exportTimeRange;
exportSession.audioMix = exportAudioMix;
// MAKE THE EXPORT SYNCHRONOUS
dispatch_semaphore_t semaphore = dispatch_semaphore_create(0);
[exportSession exportAsynchronouslyWithCompletionHandler:^{
dispatch_semaphore_signal(semaphore);
}];
dispatch_semaphore_wait(semaphore, DISPATCH_TIME_FOREVER);
if (AVAssetExportSessionStatusCompleted == exportSession.status) {
NSLog(#"AVAssetExportSessionStatusCompleted");
return trimmedAudioFilePath;
} else if (AVAssetExportSessionStatusFailed == exportSession.status) {
// a failure may happen because of an event out of your control
// for example, an interruption like a phone call comming in
// make sure and handle this case appropriately
NSLog(#"AVAssetExportSessionStatusFailed %#", exportSession.error.localizedDescription);
} else {
NSLog(#"Export Session Status: %d", exportSession.status);
}
return nil;
}
I'm recording the audio with an AVAudioRecorder, and saving it to an .aif file. I've seen some mention elsewhere of methods by which I could have it wait to start recording until the audio level reaches a threshold; that'd get me halfway there
Without adequate buffering, that would truncate the start.
I don't know of an easy way. You would have to write a new audio file after recording and analyzing it for the desired start and end points. Modifying the existing file would be straightforward if you knew the AIFF format well (not many people do) and had an easy way to read the file's sample data.
The analysis stage is pretty easy for a basic implementation -- evaluate the average power of sample data, until your threshold is exceeded. Repeat in reverse for end.

tracks in AVComposition losing time when paused

I've created an AVMutableComposition that consists of a bunch of audio tracks that start at specific times. From there, following Apple recommendations, i turned it into an AVComposition before playing it with AVPlayer.
It all works fine playing this AVPlayer item, but if I pause it and then continue, all the tracks in the composition appear to slip back about 0.2 seconds relative to each other (i.e., they bunch up). Hitting pause and continuing several times compounds the effect and the overlap is more significant (basically if I hit it enough, I will end up with all 8 tracks playing simultaneously).
if (self.player.rate > 0.0) {
//if player is playing, pause
[self.player pause];
} else {
if (self.player) {
[self.player play];
return;
}
*/CODE CREATING COMPOSITION - missed out big chunk of code relating to finding the track and retrieving its position and scale/*
NSDictionary *options = [NSDictionary dictionaryWithObject:[NSNumber numberWithBool:YES]
forKey:AVURLAssetPreferPreciseDurationAndTimingKey];
AVURLAsset *sourceAsset = [[AVURLAsset alloc] initWithURL:url options:options];
//calculate times
NSNumber *time = [soundArray1 objectAtIndex:1]; //this is the time scale - e.g. 96 or 120 etc.
double timenow = [time doubleValue];
double insertTime = (240*y);
AVMutableCompositionTrack *track =
[composition addMutableTrackWithMediaType:AVMediaTypeAudio
preferredTrackID:kCMPersistentTrackID_Invalid];
//insert the audio track from the asset into the track added to the mutable composition
AVAssetTrack *myTrack = [[sourceAsset tracksWithMediaType:AVMediaTypeAudio] objectAtIndex:0];
CMTimeRange myTrackRange = myTrack.timeRange;
NSError *error = nil;
[track insertTimeRange:myTrackRange
ofTrack:myTrack
atTime:CMTimeMake(insertTime, timenow)
error:&error];
[sourceAsset release];
}
}
AVComposition *immutableSnapshotOfMyComposition = [composition copy];
AVPlayerItem *playerItem = [AVPlayerItem playerItemWithAsset:immutableSnapshotOfMyComposition];
self.player = [[AVPlayer alloc] initWithPlayerItem:playerItem];
NSLog(#"here");
[self.player play];
Thanks
OK, this feels a little hacky, but it definitely works if anybody is stuck. If someone has a better answer, do let me know!
Basically, I just save the player.currentTime of the track when I hit pause and remake the track when i hit play, just starting from the point at which i paused it. No discernible delay, but I'd still be happier without wasting this extra processing.
Make sure you properly release your player item after you hit pause, otherwise you'll end up with a giant stack of AVPlayers!
I have a solution that is a bit less hacky but still hacky.
The solution comes from the fact that I noticed that if you seeked on the player, the latency between audio and video introduced by pausing disappeared.
Hence: just save the player.currentTime just before pausing and, player seekToTime just before playing again. It works pretty well on iOS 6, haven't tested on other versions yet.

Resources