如何在AVPlayer播放无音HLS视频时添加外部m3u8音轨？

Absolutely possible! I’ve dealt with this exact use case while building a media streaming app, so let’s break down how to sync your external HLS audio track with the video in progress.

The core idea is leveraging AVFoundation’s composition capabilities to merge the silent video stream and standalone audio stream into a unified playback resource—even while the video is already playing. Here’s a step-by-step implementation:

1. Load Video and Audio Assets Asynchronously #

First, you need to load both the video and audio AVURLAssets and verify their tracks are available. Always use async loading to avoid blocking the main thread:

// Initialize assets for your silent video and external audio
let videoURL = URL(string: "your-silent-video-m3u8-url")!
let videoAsset = AVURLAsset(url: videoURL)

let audioURL = URL(string: "your-external-audio-m3u8-url")!
let audioAsset = AVURLAsset(url: audioURL)

// Use a dispatch group to wait for both assets to load
let loadGroup = DispatchGroup()

// Load video tracks
loadGroup.enter()
videoAsset.loadValuesAsynchronously(forKeys: ["tracks"]) {
    defer { loadGroup.leave() }
    guard videoAsset.statusOfValue(forKey: "tracks", error: nil) == .loaded else {
        print("Failed to load video tracks")
        return
    }
}

// Load audio tracks
loadGroup.enter()
audioAsset.loadValuesAsynchronously(forKeys: ["tracks"]) {
    defer { loadGroup.leave() }
    guard audioAsset.statusOfValue(forKey: "tracks", error: nil) == .loaded else {
        print("Failed to load audio tracks")
        return
    }
}

// Proceed once both assets are ready
loadGroup.notify(queue: .main) {
    self.mergeVideoAndAudio(videoAsset: videoAsset, audioAsset: audioAsset)
}

2. Merge Tracks with AVMutableComposition #

Use AVMutableComposition to combine the video track from your original stream and the audio track from the external source, then switch your AVPlayer to use this new composite item:

func mergeVideoAndAudio(videoAsset: AVURLAsset, audioAsset: AVURLAsset) {
    let composition = AVMutableComposition()
    
    // Add video track to composition
    guard let sourceVideoTrack = videoAsset.tracks(withMediaType: .video).first else {
        print("No video track found in source asset")
        return
    }
    let compositionVideoTrack = composition.addMutableTrack(
        withMediaType: .video,
        preferredTrackID: kCMPersistentTrackID_Invalid
    )!
    do {
        try compositionVideoTrack.insertTimeRange(
            CMTimeRange(start: .zero, duration: videoAsset.duration),
            of: sourceVideoTrack,
            at: .zero
        )
    } catch {
        print("Failed to insert video track: \(error.localizedDescription)")
        return
    }
    
    // Add audio track to composition
    guard let sourceAudioTrack = audioAsset.tracks(withMediaType: .audio).first else {
        print("No audio track found in external asset")
        return
    }
    let compositionAudioTrack = composition.addMutableTrack(
        withMediaType: .audio,
        preferredTrackID: kCMPersistentTrackID_Invalid
    )!
    do {
        try compositionAudioTrack.insertTimeRange(
            CMTimeRange(start: .zero, duration: audioAsset.duration),
            of: sourceAudioTrack,
            at: .zero
        )
    } catch {
        print("Failed to insert audio track: \(error.localizedDescription)")
        return
    }
    
    // Create a new player item from the composition
    let compositePlayerItem = AVPlayerItem(asset: composition)
    
    // Switch to the composite item seamlessly
    if let currentPlayer = self.player {
        let wasPlaying = currentPlayer.timeControlStatus == .playing
        currentPlayer.replaceCurrentItem(with: compositePlayerItem)
        if wasPlaying {
            currentPlayer.play()
        }
    } else {
        // Initialize player if it doesn't exist yet
        self.player = AVPlayer(playerItem: compositePlayerItem)
        self.player?.play()
    }
}

3. Handle Sync & Edge Cases #

To ensure smooth playback, keep these points in mind:

• Timestamp Alignment: If your video and audio streams have mismatched timestamps, adjust the at parameter in insertTimeRange to manually offset one track to match the other.
• Duration Mismatch: If one track is longer than the other, either crop the longer track or loop the shorter one by inserting its time range multiple times into the composition.
• DRM Compatibility: If your content is DRM-protected, both the video and audio streams must use the same DRM configuration (e.g., FairPlay) for the composite asset to play.
• Error Recovery: Add fallback logic (like reverting to silent playback) if the external audio track fails to load mid-playback.

内容的提问来源于stack exchange，提问作者Nikhil Gupta