AudioKit 5：如何在Xcode/Swift中归一化.caf格式音频文件

直接基于AVFoundation实现CAF音频的音量归一化，无需依赖AudioKit，核心逻辑是读取音频样本、计算峰值、调整增益后重新导出。

完整实现代码 #

import AVFoundation

/// 归一化CAF音频文件，将最大音量调整到目标dB值
/// - Parameters:
///   - url: 源CAF文件URL
///   - targetMaxLevelDB: 目标最大音量（dB，默认-4dB）
/// - Returns: 归一化后的音频文件URL
/// - Throws: 处理过程中的错误
func normalizeCAFAudio(at url: URL, targetMaxLevelDB: Float = -4) throws -> URL {
    // 读取源音频文件
    let sourceFile = try AVAudioFile(forReading: url)
    let format = sourceFile.processingFormat
    let frameCount = UInt32(sourceFile.length)
    
    // 创建音频缓冲区存储样本数据
    guard let buffer = AVAudioPCMBuffer(pcmFormat: format, frameCapacity: frameCount) else {
        throw NSError(domain: "NormalizationError", code: 1, userInfo: [NSLocalizedDescriptionKey: "无法创建音频缓冲区"])
    }
    try sourceFile.read(into: buffer)
    
    // 计算当前音频的最大峰值（线性值）
    var maxPeak: Float = 0
    guard let channelData = buffer.floatChannelData else {
        throw NSError(domain: "NormalizationError", code: 2, userInfo: [NSLocalizedDescriptionKey: "无法获取音频通道数据"])
    }
    let channelCount = Int(format.channelCount)
    let sampleCount = Int(buffer.frameLength)
    
    for channel in 0..<channelCount {
        let samples = channelData[channel]
        for sampleIndex in 0..<sampleCount {
            let sample = abs(samples[sampleIndex])
            if sample > maxPeak {
                maxPeak = sample
            }
        }
    }
    
    // 静音文件直接返回原URL
    guard maxPeak > 0 else {
        return url
    }
    
    // 将目标dB值转换为线性振幅
    let targetMaxLinear = pow(10, targetMaxLevelDB / 20)
    // 计算增益：避免过度放大导致削波，最大增益限制为1
    let gain = min(targetMaxLinear / maxPeak, 1.0)
    
    // 应用增益到所有音频样本
    for channel in 0..<channelCount {
        let samples = channelData[channel]
        for sampleIndex in 0..<sampleCount {
            samples[sampleIndex] *= gain
        }
    }
    
    // 创建输出文件（原路径添加_normalized后缀）
    let outputFileName = url.deletingPathExtension().lastPathComponent + "_normalized.caf"
    let outputURL = url.deletingLastPathComponent().appendingPathComponent(outputFileName)
    
    // 写入归一化后的音频数据
    let outputFile = try AVAudioFile(forWriting: outputURL, settings: sourceFile.fileFormat.settings, commonFormat: format.commonFormat, interleaved: format.isInterleaved)
    try outputFile.write(from: buffer)
    
    print("归一化完成，输出路径：\(outputURL.path)")
    return outputURL
}

使用示例 #

let inputCAFURL = URL(fileURLWithPath: "/Users/xxx/Documents/recorded.caf")
do {
    let normalizedURL = try normalizeCAFAudio(at: inputCAFURL)
    // 后续处理归一化后的文件
} catch {
    print("归一化失败：\(error.localizedDescription)")
}

注意事项 #

• 上述代码一次性读取全部音频样本，适合中小体积的CAF文件；处理大文件时建议分块读取，降低内存占用
• 增益限制为≤1是为了防止原本音量高于目标值时被放大导致削波；若需要对低音量音频进行放大，可移除min(..., 1.0)的限制，但需注意监听是否出现音频失真
• 确保项目已导入AVFoundation框架，无需额外依赖第三方库

内容的提问来源于stack exchange，提问作者user2876642