【摘要】 目的分析常见的30种录音笔录制的音频,通过考察音频文件自身的数据情况,对音频属于何种设备录制进行识别。方法使用Ultra Edit、Win Hex等软件对音频的文件头和文件数据结构进行分析和比较,研究不同录音笔录制的音频文件格式、参数和数据结构之间的差异。结果通过录音文件的文件名、文件格式、文件头和文件结构能够较好地区分文件来源,其中,WMA和REC格式文件能达到100%的识别准确率,MP3格式文件的区分率为96.67%,WAV格式文件区分率为88.1%,说明该方法具有较强的区分性能。结论实验结果显示该方法使用常见工具,分析步骤简单,考察点清晰,识别率高,能够广泛应用于录音真实性鉴定实践中。更多还原

【Abstract】 Objective To establish a method for the identification of audio file sources based on data analysis.Method The audio files recorded by 30 common portable audio recorders were analyzed with decoding software,such as Ultra Edit and Win Hex.The file head and structure of audio files recorded by different recorders were analyzed comparatively to study the difference in the formats,parameters and structures of these files.Results The audio files can be well classified by the combination of file name,format,file head and structure.The recognition rates were 100% for WMA and REC files,96.67%for MP3 files,and 88.1% for WAV files.Conclusion The established method is simple and efficient,as it acquires high recognition rate using common software.It can be widely used in the forensic practice of audio authentication.更多还原