You need to enable JavaScript to run this app. AWS Polly处理含特殊字符文本时字节偏移转字符偏移问题咨询
优惠活动
大模型
产品
解决方案
定价
更多

AWS Polly处理含特殊字符文本时字节偏移转字符偏移问题咨询

AWS Polly字节偏移转字符偏移实现文本准确高亮的方案

问题根源

AWS Polly返回的start、end字段为UTF-8字节偏移量,而Swift中字符串的字符计数、文本高亮所用的NSRange分别基于Unicode字符、UTF-16编码,遇到多字节Unicode字符(比如示例中的ö占2个UTF-8字节)时直接使用字节偏移必然出现错位。

解决方案

步骤1:新增字符串偏移转换工具

给Swift的String加扩展,实现UTF-8字节偏移到字符索引的转换:

extension String {
    /// 输入UTF8字节偏移,返回对应的字符索引
    func characterIndex(from byteOffset: Int) -> String.Index? {
        guard byteOffset >= 0, 
              let utf8Index = utf8.index(utf8.startIndex, offsetBy: byteOffset, limitedBy: utf8.endIndex) else {
            return nil
        }
        // 把UTF8视图的索引转成普通字符索引
        return String.Index(utf8Index, within: self)
    }
}

步骤2:改造高亮逻辑

核心是先通过原始输入文本(传给Polly的完整未修改文本,假设变量名为originalInputText)把字节偏移转成字符维度的范围,再转为控件可用的NSRange:

// 1. 把Polly返回的字节偏移转成字符索引
guard let startCharIndex = originalInputText.characterIndex(from: Int(pollyVoiceSentence.start)),
      let endCharIndex = originalInputText.characterIndex(from: Int(pollyVoiceSentence.end)) else {
    // 偏移异常的兜底处理
    return
}
// 2. 构造字符维度的Range
let charRange = startCharIndex..<endCharIndex
// 3. 转成NSRange用于UILabel等控件的属性文本高亮
let voiceRange = NSRange(charRange, in: originalInputText)

print("RANGE: \(voiceRange) - Word: \(pollyVoiceSentence.value)")

可选优化:解析阶段预转换偏移

如果不想每次高亮都重复转换,可以在解析Polly返回结果时直接把字节偏移转成字符偏移存入模型,需要先给PollyVoiceSentence新增charStart、charEnd两个Int属性:

builder.continueOnSuccessWith { (awsTask: AWSTask<NSURL>) -> Any? in
    if builder.error == nil {
        if let url = awsTask.result {
            do {
                let txtData = try Data(contentsOf: url as URL)
                if let txtString = String(data: txtData, encoding: .utf8) {
                    let lines = txtString.components(separatedBy: .newlines)
                    for line in lines {
                        let jsonData = Data(line.utf8)
                        let pollyVoiceSentence = try JSONDecoder().decode(PollyVoiceSentence.self, from: jsonData)
                        // 预转换为字符偏移
                        if let startIdx = originalInputText.characterIndex(from: Int(pollyVoiceSentence.start)),
                           let endIdx = originalInputText.characterIndex(from: Int(pollyVoiceSentence.end)) {
                            pollyVoiceSentence.charStart = originalInputText.distance(from: originalInputText.startIndex, to: startIdx)
                            pollyVoiceSentence.charEnd = originalInputText.distance(from: originalInputText.startIndex, to: endIdx)
                        }
                        voiceSentences.append(pollyVoiceSentence)
                    }
                }
            } catch {
                print("Could not parse TXT file")
            }
        }
    } else {
        print("ParseJSON: \(builder.error!)")
    }
    completionHandler(voiceSentences)
    return nil
}

注意事项

  • 必须严格保留传给Polly的原始输入文本,不能做任何修改、裁剪、转义操作,否则偏移会完全错位
  • 转换方法兼容所有Unicode字符,包括特殊字符、emoji、东亚文字等场景

内容的提问来源于stack exchange,提问作者sametbilgi

相关产品推荐
方舟 Agent Plan

超全模态模型 × Harness 升级,最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列、Doubao-Seedream-5.0-pro、Kimi-K3 (部分), 限时 9.9 元起

最近更新时间:2026.10.17 11:15:04