Datasets:

Hezep
/

AudioMarathon

@@ -1,247 +0,0 @@
-import os
-import sys
-import soundfile as sf
-import numpy as np
-from pathlib import Path
-import argparse
-from collections import defaultdict
-import traceback
-def get_audio_duration(audio_path):
-    """获取音频文件的时长（秒）"""
-    try:
-        # 使用soundfile获取音频信息
-        info = sf.info(audio_path)
-        duration = info.duration
-        return duration, info.samplerate, info.channels
-    except Exception as e:
-        print(f"警告: 无法读取音频文件 {audio_path}: {e}")
-        return None, None, None
-def analyze_audio_folder(folder_path, extensions=None):
-    """分析文件夹中所有音频文件的时长"""
-    if extensions is None:
-        extensions = ['.wav', '.mp3', '.flac', '.ogg', '.m4a', '.aac', '.wma']
-    # 确保扩展名都是小写
-    extensions = [ext.lower() for ext in extensions]
-    folder_path = Path(folder_path)
-    if not folder_path.exists():
-        print(f"错误: 文件夹不存在: {folder_path}")
-        return None
-    print(f"正在分析文件夹: {folder_path}")
-    print(f"支持的音频格式: {', '.join(extensions)}")
-    print("-" * 50)
-    # 收集所有音频文件
-    audio_files = []
-    for ext in extensions:
-        # 递归查找所有音频文件
-        pattern = f"**/*{ext}"
-        files = list(folder_path.glob(pattern))
-        audio_files.extend(files)
-    if not audio_files:
-        print("未找到任何音频文件")
-        return None
-    print(f"找到 {len(audio_files)} 个音频文件")
-    # 分析每个音频文件
-    durations = []
-    sample_rates = []
-    channels_info = []
-    failed_files = []
-    format_stats = defaultdict(list)
-    for i, audio_file in enumerate(audio_files):
-        if (i + 1) % 100 == 0:
-            print(f"进度: {i+1}/{len(audio_files)} ({(i+1)/len(audio_files)*100:.1f}%)")
-        duration, sample_rate, channels = get_audio_duration(audio_file)
-        if duration is not None:
-            durations.append(duration)
-            sample_rates.append(sample_rate)
-            channels_info.append(channels)
-            # 按文件格式分类统计
-            file_ext = audio_file.suffix.lower()
-            format_stats[file_ext].append(duration)
-        else:
-            failed_files.append(audio_file)
-    if not durations:
-        print("没有成功读取任何音频文件")
-        return None
-    # 计算统计信息
-    durations = np.array(durations)
-    sample_rates = np.array(sample_rates)
-    total_duration = np.sum(durations)
-    avg_duration = np.mean(durations)
-    max_duration = np.max(durations)
-    min_duration = np.min(durations)
-    median_duration = np.median(durations)
-    std_duration = np.std(durations)
-    # 找出最长和最短的文件
-    max_idx = np.argmax(durations)
-    min_idx = np.argmin(durations)
-    max_file = audio_files[max_idx] if max_idx < len(audio_files) else "未知"
-    min_file = audio_files[min_idx] if min_idx < len(audio_files) else "未知"
-    # 输出统计结果
-    print("\n" + "=" * 60)
-    print("音频文件时长统计结果")
-    print("=" * 60)
-    print(f"总文件数: {len(audio_files)}")
-    print(f"成功读取: {len(durations)}")
-    print(f"读取失败: {len(failed_files)}")
-    print(f"\n时长统计:")
-    print(f"  总时长: {total_duration:.2f} 秒 ({total_duration/60:.2f} 分钟, {total_duration/3600:.2f} 小时)")
-    print(f"  平均时长: {avg_duration:.2f} 秒")
-    print(f"  最大时长: {max_duration:.2f} 秒")
-    print(f"  最小时长: {min_duration:.2f} 秒")
-    print(f"  中位数时长: {median_duration:.2f} 秒")
-    print(f"  标准差: {std_duration:.2f} 秒")
-    print(f"\n文件信息:")
-    print(f"  最长文件: {max_file.name} ({max_duration:.2f}秒)")
-    print(f"  最短文件: {min_file.name} ({min_duration:.2f}秒)")
-    # 采样率统计
-    unique_sample_rates = np.unique(sample_rates)
-    print(f"\n采样率统计:")
-    for sr in unique_sample_rates:
-        count = np.sum(sample_rates == sr)
-        print(f"  {sr} Hz: {count} 个文件")
-    # 声道数统计
-    unique_channels = np.unique(channels_info)
-    print(f"\n声道数统计:")
-    for ch in unique_channels:
-        count = np.sum(np.array(channels_info) == ch)
-        print(f"  {ch} 声道: {count} 个文件")
-    # 按文件格式统计
-    print(f"\n按文件格式统计:")
-    for ext, ext_durations in format_stats.items():
-        ext_durations = np.array(ext_durations)
-        print(f"  {ext}: {len(ext_durations)} 个文件")
-        print(f"    平均时长: {np.mean(ext_durations):.2f} 秒")
-        print(f"    最大时长: {np.max(ext_durations):.2f} 秒")
-        print(f"    最小时长: {np.min(ext_durations):.2f} 秒")
-    # 时长分布统计
-    print(f"\n时长分布统计:")
-    ranges = [(0, 10), (10, 30), (30, 60), (60, 180), (180, 600), (600, float('inf'))]
-    range_names = ["0-10秒", "10-30秒", "30-60秒", "1-3分钟", "3-10分钟", "10分钟以上"]
-    for (start, end), name in zip(ranges, range_names):
-        if end == float('inf'):
-            count = np.sum(durations >= start)
-        else:
-            count = np.sum((durations >= start) & (durations < end))
-        percentage = count / len(durations) * 100
-        print(f"  {name}: {count} 个文件 ({percentage:.1f}%)")
-    # 显示失败的文件
-    if failed_files:
-        print(f"\n读取失败的文件 ({len(failed_files)} 个):")
-        for failed_file in failed_files[:10]:  # 只显示前10个
-            print(f"  {failed_file}")
-        if len(failed_files) > 10:
-            print(f"  ... 还有 {len(failed_files) - 10} 个文件")
-    print("\n" + "=" * 60)
-    return {
-        "total_files": len(audio_files),
-        "successful_files": len(durations),
-        "failed_files": len(failed_files),
-        "total_duration": total_duration,
-        "avg_duration": avg_duration,
-        "max_duration": max_duration,
-        "min_duration": min_duration,
-        "median_duration": median_duration,
-        "std_duration": std_duration,
-        "max_file": str(max_file),
-        "min_file": str(min_file),
-        "durations": durations.tolist(),
-        "sample_rates": sample_rates.tolist(),
-        "channels": channels_info,
-        "format_stats": {k: np.array(v).tolist() for k, v in format_stats.items()},
-        "failed_files": [str(f) for f in failed_files]
-    }
-def main():
-    parser = argparse.ArgumentParser(description='分析文件夹中音频文件的时长统计')
-    parser.add_argument('folder', help='要分析的文件夹路径')
-    parser.add_argument('--extensions', nargs='*',
-                       default=['.wav', '.mp3', '.flac', '.ogg', '.m4a', '.aac', '.wma'],
-                       help='支持的音频文件扩展名')
-    parser.add_argument('--output', '-o', help='输出JSON结果文件路径')
-    args = parser.parse_args()
-    try:
-        # 分析音频文件夹
-        results = analyze_audio_folder(args.folder, args.extensions)
-        if results and args.output:
-            import json
-            with open(args.output, 'w', encoding='utf-8') as f:
-                json.dump(results, f, ensure_ascii=False, indent=2)
-            print(f"\n详细结果已保存到: {args.output}")
-    except Exception as e:
-        print(f"程序执行出错: {e}")
-        traceback.print_exc()
-        sys.exit(1)
-if __name__ == "__main__":
-    # 如果没有命令行参数，提供交互式使用
-    if len(sys.argv) == 1:
-        print("音频文件时长分析工具")
-        print("用法1: python timing.py <文件夹路径>")
-        print("用法2: python timing.py <文件夹路径> --output results.json")
-        print("用法3: python timing.py <文件夹路径> --extensions .wav .mp3")
-        print("\n交互式使用:")
-        folder_path = input("请输入要分析的文件夹路径: ").strip()
-        if not folder_path:
-            print("错误: 请提供有效的文件夹路径")
-            sys.exit(1)
-        # 使用默认设置分析
-        try:
-            results = analyze_audio_folder(folder_path)
-            # 询问是否保存结果
-            save_result = input("\n是否保存详细结果到JSON文件? (y/n): ").strip().lower()
-            if save_result in ['y', 'yes']:
-                output_file = input("请输入输出文件路径 (默认: audio_analysis_results.json): ").strip()
-                if not output_file:
-                    output_file = "audio_analysis_results.json"
-                import json
-                with open(output_file, 'w', encoding='utf-8') as f:
-                    json.dump(results, f, ensure_ascii=False, indent=2)
-                print(f"详细结果已保存到: {output_file}")
-        except Exception as e:
-            print(f"程序执行出错: {e}")
-            traceback.print_exc()
-            sys.exit(1)
-    else:
-        main()