中文说明：读取指定文件夹下的所有mp3文件，转换成语谱图，并保存为二进制文件，便于后续进行语言识别等处理。附带测试样本音频。
English Description：
&Lt spanstyle = & quot; color: ᦇ 666666; font family: & quot; background color: ᦇ ffffff; & quot; & gt; read all the MP3 files in the specified folder, convert the idiom spectrum and save it as a binary file for subsequent language recognition and other processing. Test sample audio is attached.