[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-musicinformationretrieval--musicinformationretrieval.com":3,"tool-musicinformationretrieval--musicinformationretrieval.com":61},[4,18,26,36,44,52],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",141543,2,"2026-04-06T11:32:54",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107888,"2026-04-06T11:32:50",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":10,"last_commit_at":50,"category_tags":51,"status":17},4487,"LLMs-from-scratch","rasbt\u002FLLMs-from-scratch","LLMs-from-scratch 是一个基于 PyTorch 的开源教育项目，旨在引导用户从零开始一步步构建一个类似 ChatGPT 的大型语言模型（LLM）。它不仅是同名技术著作的官方代码库，更提供了一套完整的实践方案，涵盖模型开发、预训练及微调的全过程。\n\n该项目主要解决了大模型领域“黑盒化”的学习痛点。许多开发者虽能调用现成模型，却难以深入理解其内部架构与训练机制。通过亲手编写每一行核心代码，用户能够透彻掌握 Transformer 架构、注意力机制等关键原理，从而真正理解大模型是如何“思考”的。此外，项目还包含了加载大型预训练权重进行微调的代码，帮助用户将理论知识延伸至实际应用。\n\nLLMs-from-scratch 特别适合希望深入底层原理的 AI 开发者、研究人员以及计算机专业的学生。对于不满足于仅使用 API，而是渴望探究模型构建细节的技术人员而言，这是极佳的学习资源。其独特的技术亮点在于“循序渐进”的教学设计：将复杂的系统工程拆解为清晰的步骤，配合详细的图表与示例，让构建一个虽小但功能完备的大模型变得触手可及。无论你是想夯实理论基础，还是为未来研发更大规模的模型做准备",90106,"2026-04-06T11:19:32",[35,15,13,14],{"id":53,"name":54,"github_repo":55,"description_zh":56,"stars":57,"difficulty_score":10,"last_commit_at":58,"category_tags":59,"status":17},4292,"Deep-Live-Cam","hacksider\u002FDeep-Live-Cam","Deep-Live-Cam 是一款专注于实时换脸与视频生成的开源工具，用户仅需一张静态照片，即可通过“一键操作”实现摄像头画面的即时变脸或制作深度伪造视频。它有效解决了传统换脸技术流程繁琐、对硬件配置要求极高以及难以实时预览的痛点，让高质量的数字内容创作变得触手可及。\n\n这款工具不仅适合开发者和技术研究人员探索算法边界，更因其极简的操作逻辑（仅需三步：选脸、选摄像头、启动），广泛适用于普通用户、内容创作者、设计师及直播主播。无论是为了动画角色定制、服装展示模特替换，还是制作趣味短视频和直播互动，Deep-Live-Cam 都能提供流畅的支持。\n\n其核心技术亮点在于强大的实时处理能力，支持口型遮罩（Mouth Mask）以保留使用者原始的嘴部动作，确保表情自然精准；同时具备“人脸映射”功能，可同时对画面中的多个主体应用不同面孔。此外，项目内置了严格的内容安全过滤机制，自动拦截涉及裸露、暴力等不当素材，并倡导用户在获得授权及明确标注的前提下合规使用，体现了技术发展与伦理责任的平衡。",88924,"2026-04-06T03:28:53",[14,15,13,60],"视频",{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":67,"readme_en":68,"readme_zh":69,"quickstart_zh":70,"use_case_zh":71,"hero_image_url":72,"owner_login":73,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":75,"owner_location":75,"owner_email":75,"owner_twitter":75,"owner_website":75,"owner_url":76,"languages":77,"stars":98,"forks":99,"last_commit_at":100,"license":101,"difficulty_score":32,"env_os":102,"env_gpu":103,"env_ram":103,"env_deps":104,"category_tags":114,"github_topics":117,"view_count":32,"oss_zip_url":75,"oss_zip_packed_at":75,"status":17,"created_at":124,"updated_at":125,"faqs":126,"releases":157},4434,"musicinformationretrieval\u002Fmusicinformationretrieval.com","musicinformationretrieval.com","Instructional notebooks on music information retrieval.","musicinformationretrieval.com 是一个专注于音乐信息检索（MIR）领域的开源教学平台，旨在通过一系列交互式的 Jupyter Notebook 教程，帮助用户系统掌握从基础理论到代码实现的完整技能树。它主要解决了初学者在进入 MIR 领域时面临的门槛高、资源分散以及理论与实践脱节的问题，将复杂的声学概念转化为可运行、可视化的代码示例。\n\n这套资源非常适合计算机音乐研究者、音频算法开发者、数据科学家以及希望深入理解音频处理技术的学生使用。无论是需要快速上手 Python 音频库的新手，还是寻求特定特征提取方法参考的专业人士，都能从中获益。\n\n其独特的技术亮点在于“边学边做”的沉浸式体验：内容涵盖乐谱与符号表示、调律系统等音乐理论基础，并深入讲解傅里叶变换、短时傅里叶变换（STFT）、恒 Q 变换（CQT）及色度特征等核心信号分析技术。更难得的是，项目引入了“声音化（Sonification）”演示，让用户能直接听到音频特征的变化，从而直观理解抽象的数学原理。配合对 NumPy、SciPy、SoX 及 ffmpeg 等工具的实战讲解，musicinformati","musicinformationretrieval.com 是一个专注于音乐信息检索（MIR）领域的开源教学平台，旨在通过一系列交互式的 Jupyter Notebook 教程，帮助用户系统掌握从基础理论到代码实现的完整技能树。它主要解决了初学者在进入 MIR 领域时面临的门槛高、资源分散以及理论与实践脱节的问题，将复杂的声学概念转化为可运行、可视化的代码示例。\n\n这套资源非常适合计算机音乐研究者、音频算法开发者、数据科学家以及希望深入理解音频处理技术的学生使用。无论是需要快速上手 Python 音频库的新手，还是寻求特定特征提取方法参考的专业人士，都能从中获益。\n\n其独特的技术亮点在于“边学边做”的沉浸式体验：内容涵盖乐谱与符号表示、调律系统等音乐理论基础，并深入讲解傅里叶变换、短时傅里叶变换（STFT）、恒 Q 变换（CQT）及色度特征等核心信号分析技术。更难得的是，项目引入了“声音化（Sonification）”演示，让用户能直接听到音频特征的变化，从而直观理解抽象的数学原理。配合对 NumPy、SciPy、SoX 及 ffmpeg 等工具的实战讲解，musicinformationretrieval.com 成为了连接音乐理论与工程实践的理想桥梁。","musicinformationretrieval.com\n=============================\n\nIntroduction\n------------\n1.  [About This Site](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fabout.ipynb)\n2.  [What is MIR?](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fwhy_mir.ipynb)\n3.  [Python Basics and Dependencies](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fpython_basics.ipynb)\n4.  [Jupyter Basics](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fget_good_at_ipython.ipynb)\n5.  [Jupyter Audio Basics](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fipython_audio.ipynb)\n6.  [SoX and ffmpeg](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fsox_and_ffmpeg.ipynb)\n7.  [NumPy and SciPy Basics](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fnumpy_basics.ipynb)\n\nMusic Representations\n---------------------\n1.  [Sheet Music Representations](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fsheet_music_representations.ipynb)\n2.  [Symbolic Representations](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fsymbolic_representations.ipynb)\n3.  [Audio Representation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Faudio_representation.ipynb)\n4.  [Tuning Systems](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Ftuning_systems.ipynb)\n5.  [MIDI Note to Frequency Conversion Table](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fmidi_conversion_table.ipynb)\n6.  [Understanding Audio Features through Sonification](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Ffeature_sonification.ipynb)\n\nSignal Analysis and Feature Extraction\n--------------------------------------\n1.  [Basic Feature Extraction](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fbasic_feature_extraction.ipynb)\n2.  [Segmentation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fsegmentation.ipynb)\n3.  [Energy and RMSE](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fenergy.ipynb)\n4.  [Zero Crossing Rate](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fzcr.ipynb)\n5.  [Fourier Transform](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Ffourier_transform.ipynb)\n6.  [Short-time Fourier Transform and Spectrogram](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fstft.ipynb)\n7.  [Constant-Q Transform and Chroma](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fchroma.ipynb)\n8.  [Video: Chroma Features](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fvideo_chroma.ipynb)\n9.  [Magnitude Scaling](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fmagnitude_scaling.ipynb)\n10. [Spectral Features](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fspectral_features.ipynb)\n11. [Autocorrelation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fautocorrelation.ipynb)\n12. [Pitch Transcription Exercise](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fpitch_transcription_exercise.ipynb)\n\nRhythm, Tempo, and Beat Tracking\n--------------------------------\n1. [Novelty Functions](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fnovelty_functions.ipynb)\n2. [Peak Picking](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fpeak_picking.ipynb)\n3. [Onset Detection](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fonset_detection.ipynb)\n4. [Onset-based Segmentation with Backtracking](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fonset_segmentation.ipynb)\n5. [Tempo Estimation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Ftempo_estimation.ipynb)\n6. [Beat Tracking](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fbeat_tracking.ipynb)\n7. [Video: Tempo and Beat Tracking](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fvideo_tempo_beat_tracking.ipynb)\n8. [Drum Transcription using ADTLib](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fadtlib.ipynb)\n\nMachine Learning\n----------------\n1. [K-Means Clustering](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fkmeans.ipynb)\n2. [Exercise: Unsupervised Instrument Classification using K-Means](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fkmeans_instrument_classification_exercise.ipynb)\n3. [Neural Networks](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fneural_networks.ipynb)\n4. [Genre Recognition](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fgenre_recognition.ipynb)\n5. [Exercise: Genre Recognition](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fexercise_genre_recognition.ipynb)\n\nEvaluation\n----------\n1. [Introduction to \u003Ccode>mir_eval\u003C\u002Fcode>](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fintro_mir_eval.ipynb)\n2. [Onset Detection](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_onset.ipynb)\n3. [Beat Tracking](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_beat.ipynb)\n4. [Chord Estimation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_chord.ipynb)\n\nMusic Synchronization\n---------------------\n1. [Dynamic Programming](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdp.ipynb)\n2. [Longest Common Subsequence](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Flcs.ipynb)\n3. [Dynamic Time Warping](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdtw.ipynb)\n4. [Dynamic Time Warping Example](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdtw_example.ipynb)\n\nMusic Structure Analysis\n------------------------\n1. [Mel-Frequency Cepstral Coefficients](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F8_music_structure_analysis%2Fmfcc.ipynb)\n\nContent-Based Audio Retrieval\n-----------------------------\n1. [Locality Sensitive Hashing](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F9_retrieval%2Flsh_fingerprinting.ipynb)\n\nMusically Informed Audio Decomposition\n--------------------------------------\n1. [Principal Component Analysis](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fpca.ipynb)\n2. [Nonnegative Matrix Factorization](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fnmf.ipynb)\n3. [NMF Audio Mosaicing](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fnmf_audio_mosaic.ipynb)\n4. [Harmonic-Percussive Source Separation](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fhpss.ipynb)\n\nJust For Fun\n------------\n1. [Real-time Spectrogram](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F11_fun%2Frealtime_spectrogram.ipynb)\n2. [THX Logo Theme](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F11_fun%2Fthx_logo_theme.ipynb)\n\nAbout\n-----\n1. [About the book \u003Cem>Fundamentals of Music Processing\u003C\u002Fem>](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2Fx1_about%2Fabout_fmp.ipynb)\n2. [About the CCRMA Workshop on Music Information Retrieval](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2Fx1_about%2Fabout_ccrma_workshop.ipynb)\n\n---\n\nThis repository contains instructional notebooks related to music information retrieval (MIR). Inside these notebooks are Python code snippets that illustrate basic MIR systems. You can actually execute the code from inside the notebook.\n\nThe notebooks run on [`binder`](https:\u002F\u002Fmybinder.org) to enable reproducible results in a consistent computational environment.\n\n\nUpdates\n-------\n\n2025 October 22: Hey! I'm [Huw Cheston](https:\u002F\u002Fhuwcheston.github.io) and I'm excited to be working with Iran on modernising this site as part of the [EPSRC](https:\u002F\u002Fwww.ukri.org\u002Fcouncils\u002Fepsrc\u002F) \"AI Skills Through Music\" project. I completed my PhD in MIR at the [University of Cambridge](https:\u002F\u002Fcms.mus.cam.ac.uk\u002F) and now work at the [Centre for Digital Music, Queen Mary University](https:\u002F\u002Fwww.c4dm.eecs.qmul.ac.uk\u002F). Feel free to reach out and say hi! `h \\\u003Cdot\\> cheston \\\u003Cat\\> qmul \\\u003Cdot\\> ac \\\u003Cdot\\> uk`.\n\n2023 October 09: Hello everyone! I'm [Iran R. Roman](https:\u002F\u002Firanroman.github.io) and I'm honored to be [the new administrator](https:\u002F\u002Firanroman.github.io\u002F2023\u002F10\u002F09\u002Fother-list-01.html) of [musicinformationretrieval.com](https:\u002F\u002Fmusicinformationretrieval.com). Feel free to reach out and say hi. I would love to hear from you. iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu\n\n2022 April 22: It's 2022, and Colab seems to be much more popular and usable than it was a few years ago. **You can help me migrate musicinformationretrieval.com to Colab.** Edit a Colab notebook, and submit a pull request. Ping iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu to let me know.\n\nContributions\n-------------\n\nYour contributions are welcome! You can contribute in two ways:\n\n1. Submit an issue. Click on \"[Issues](https:\u002F\u002Fgithub.com\u002Firanroman\u002Fmusicinformationretrieval.com\u002Fissues)\" in the right navigation bar, then \"New Issue\".  Issues can include Python bugs, spelling mistakes, broken links, requests for new content, and more.\n\n2. Submit changes to source code or documentation. [Fork this repo](https:\u002F\u002Fhelp.github.com\u002Farticles\u002Ffork-a-repo), make edits to your fork, then [submit a pull request](https:\u002F\u002Fhelp.github.com\u002Farticles\u002Fusing-pull-requests). `gh-pages` is the default branch for this repo. Try to follow the style conventions in the existing notebooks. Ping iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu to let me know you submitted a pull request.\n","musicinformationretrieval.com\n=============================\n\n简介\n----\n1.  [关于本站](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fabout.ipynb)\n2.  [什么是MIR？](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fwhy_mir.ipynb)\n3.  [Python基础与依赖](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fpython_basics.ipynb)\n4.  [Jupyter基础](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fget_good_at_ipython.ipynb)\n5.  [Jupyter音频基础](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fipython_audio.ipynb)\n6.  [SoX与ffmpeg](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fsox_and_ffmpeg.ipynb)\n7.  [NumPy与SciPy基础](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F1_introduction%2Fnumpy_basics.ipynb)\n\n音乐表示法\n-----------\n1.  [乐谱表示法](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fsheet_music_representations.ipynb)\n2.  [符号化表示法](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fsymbolic_representations.ipynb)\n3.  [音频表示法](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Faudio_representation.ipynb)\n4.  [调音系统](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Ftuning_systems.ipynb)\n5.  [MIDI音符到频率转换表](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Fmidi_conversion_table.ipynb)\n6.  [通过声音化理解音频特征](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F2_music_representations%2Ffeature_sonification.ipynb)\n\n信号分析与特征提取\n------------------\n1.  [基础特征提取](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fbasic_feature_extraction.ipynb)\n2.  [分段](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fsegmentation.ipynb)\n3.  [能量与均方根误差](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fenergy.ipynb)\n4.  [过零率](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fzcr.ipynb)\n5.  [傅里叶变换](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Ffourier_transform.ipynb)\n6.  [短时傅里叶变换与频谱图](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fstft.ipynb)\n7.  [恒Q变换与色度特征](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fchroma.ipynb)\n8.  [视频：色度特征](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fvideo_chroma.ipynb)\n9.  [幅度归一化](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fmagnitude_scaling.ipynb)\n10. [频谱特征](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fspectral_features.ipynb)\n11. [自相关](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fautocorrelation.ipynb)\n12. [音高转录练习](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F3_signal_analysis%2Fpitch_transcription_exercise.ipynb)\n\n节奏、速度与节拍跟踪\n--------------------\n1.  [新颖性函数](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fnovelty_functions.ipynb)\n2.  [峰值检测](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fpeak_picking.ipynb)\n3.  [起始点检测](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fonset_detection.ipynb)\n4.  [基于起始点的回溯式分段](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fonset_segmentation.ipynb)\n5.  [速度估计](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Ftempo_estimation.ipynb)\n6.  [节拍跟踪](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fbeat_tracking.ipynb)\n7.  [视频：速度与节拍跟踪](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fvideo_tempo_beat_tracking.ipynb)\n8.  [使用ADTLib进行鼓乐转录](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F4_rhythm_tempo_beat%2Fadtlib.ipynb)\n\n机器学习\n----------------\n1. [K均值聚类](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fkmeans.ipynb)\n2. [练习：使用K均值进行无监督乐器分类](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fkmeans_instrument_classification_exercise.ipynb)\n3. [神经网络](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fneural_networks.ipynb)\n4. [流派识别](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fgenre_recognition.ipynb)\n5. [练习：流派识别](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F5_machine_learning%2Fexercise_genre_recognition.ipynb)\n\n评估\n----------\n1. [\u003Ccode>mir_eval\u003C\u002Fcode>简介](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fintro_mir_eval.ipynb)\n2. [节拍检测](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_onset.ipynb)\n3. [节拍跟踪](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_beat.ipynb)\n4. [和弦估计](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F6_evaluation%2Fevaluation_chord.ipynb)\n\n音乐同步\n---------------------\n1. [动态规划](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdp.ipynb)\n2. [最长公共子序列](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Flcs.ipynb)\n3. [动态时间规整](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdtw.ipynb)\n4. [动态时间规整示例](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F7_music_synchronization%2Fdtw_example.ipynb)\n\n音乐结构分析\n------------------------\n1. [梅尔频率倒谱系数](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F8_music_structure_analysis%2Fmfcc.ipynb)\n\n基于内容的音频检索\n-----------------------------\n1. [局部敏感哈希](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F9_retrieval%2Flsh_fingerprinting.ipynb)\n\n音乐信息驱动的音频分解\n--------------------------------------\n1. [主成分分析](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fpca.ipynb)\n2. [非负矩阵分解](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fnmf.ipynb)\n3. [NMF音频马赛克](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fnmf_audio_mosaic.ipynb)\n4. [谐波-打击乐源分离](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F10_decomposition%2Fhpss.ipynb)\n\n纯粹娱乐\n------------\n1. [实时频谱图](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F11_fun%2Frealtime_spectrogram.ipynb)\n2. [THX标志主题曲](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2F11_fun%2Fthx_logo_theme.ipynb)\n\n关于\n-----\n1. [《音乐处理基础》一书简介](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2Fx1_about%2Fabout_fmp.ipynb)\n2. [CCRMA音乐信息检索研讨会简介](https:\u002F\u002Fmybinder.org\u002Fv2\u002Fgh\u002FHuwCheston\u002Fmusicinformationretrieval.com\u002FHEAD?urlpath=%2Fdoc%2Ftree%2Fcontent%2Fx1_about%2Fabout_ccrma_workshop.ipynb)\n\n---\n\n本仓库包含与音乐信息检索（MIR）相关的教学笔记本。这些笔记本中包含了用于演示基本MIR系统的Python代码片段，您可以在笔记本内部直接运行这些代码。\n\n这些笔记本运行在[`binder`](https:\u002F\u002Fmybinder.org)平台上，以确保在一致的计算环境中获得可重复的结果。\n\n\n更新\n-------\n\n2025年10月22日：大家好！我是[Huw Cheston](https:\u002F\u002Fhuwcheston.github.io)，很高兴能与Iran合作，作为[EPSRC](https:\u002F\u002Fwww.ukri.org\u002Fcouncils\u002Fepsrc\u002F)“通过音乐培养AI技能”项目的一部分，共同推进本网站的现代化改造。我在[剑桥大学](https:\u002F\u002Fcms.mus.cam.ac.uk\u002F)完成了MIR方向的博士学位，目前就职于[女王玛丽大学数字音乐中心](https:\u002F\u002Fwww.c4dm.eecs.qmul.ac.uk\u002F)。欢迎随时联系我打招呼！`h \\\u003Cdot\\> cheston \\\u003Cat\\> qmul \\\u003Cdot\\> ac \\\u003Cdot\\> uk`。\n\n2023年10月9日：大家好！我是[Iran R. Roman](https:\u002F\u002Firanroman.github.io)，很荣幸成为[musicinformationretrieval.com](https:\u002F\u002Fmusicinformationretrieval.com)的[新管理员](https:\u002F\u002Firanroman.github.io\u002F2023\u002F10\u002F09\u002Fother-list-01.html)。欢迎大家随时联系我打招呼，我很期待与各位交流！iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu\n\n2022年4月22日：现在是2022年，Colab似乎比几年前更加流行且易于使用。**您可以帮助我将musicinformationretrieval.com迁移到Colab平台。** 编辑一个Colab笔记本，并提交拉取请求。请告知我：iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu。\n\n贡献\n-------------\n\n我们欢迎您的贡献！您可以通过两种方式参与：\n\n1. 提交问题。点击右侧导航栏中的“[Issues](https:\u002F\u002Fgithub.com\u002Firanroman\u002Fmusicinformationretrieval.com\u002Fissues)”，然后选择“New Issue”。问题可以包括Python代码错误、拼写错误、失效链接、对新内容的需求等。\n\n2. 提交对源代码或文档的更改。[fork此仓库](https:\u002F\u002Fhelp.github.com\u002Farticles\u002Ffork-a-repo)，在您的分支上进行编辑，然后[提交pull request](https:\u002F\u002Fhelp.github.com\u002Farticles\u002Fusing-pull-requests)。`gh-pages`是该仓库的默认分支。请尽量遵循现有笔记本中的风格规范。如果您提交了拉取请求，请告知我：iran \\\u003Cat\\> ccrma \\\u003Cdot\\> stanford \\\u003Cdot\\> edu。","# musicinformationretrieval.com 快速上手指南\n\n`musicinformationretrieval.com` 是一个开源的音乐信息检索（MIR）教学项目，包含一系列基于 Python 和 Jupyter Notebook 的交互式教程，涵盖从基础信号处理到机器学习在音乐分析中的应用。\n\n## 环境准备\n\n本项目设计为在云端或本地环境中运行，推荐优先使用云端环境以避免复杂的依赖配置。\n\n### 系统要求\n- **操作系统**：Windows, macOS, 或 Linux\n- **浏览器**：现代浏览器（Chrome, Firefox, Edge 等）\n- **内存**：建议 4GB 以上（本地运行时）\n\n### 前置依赖\n若选择**本地运行**，需安装以下工具：\n- Python 3.8+\n- Jupyter Notebook \u002F JupyterLab\n- 核心库：`numpy`, `scipy`, `librosa`, `matplotlib`, `mir_eval`, `soundfile`\n- 外部工具：`ffmpeg`, `sox` (用于音频处理)\n\n> **注意**：本项目官方推荐使用 **[Binder](https:\u002F\u002Fmybinder.org)** 在线运行，无需本地安装任何依赖即可直接体验所有教程。\n\n## 安装步骤\n\n### 方案一：在线运行（推荐）\n无需安装，直接点击仓库中的链接即可在浏览器中运行代码。所有笔记本均配置了可复现的计算环境。\n\n1. 访问项目 GitHub 页面：[HuwCheston\u002Fmusicinformationretrieval.com](https:\u002F\u002Fgithub.com\u002FHuwCheston\u002Fmusicinformationretrieval.com)\n2. 点击任意 `.ipynb` 文件链接（链接格式通常为 `https:\u002F\u002Fmybinder.org\u002F...`）。\n3. 等待 Binder 构建环境（首次加载约需 1-3 分钟），即可开始交互。\n\n### 方案二：本地安装\n若需离线开发或修改代码，请按以下步骤操作：\n\n1. **克隆仓库**\n   ```bash\n   git clone https:\u002F\u002Fgithub.com\u002FHuwCheston\u002Fmusicinformationretrieval.com.git\n   cd musicinformationretrieval.com\n   ```\n\n2. **安装系统级音频工具**\n   - **Ubuntu\u002FDebian**:\n     ```bash\n     sudo apt-get update\n     sudo apt-get install ffmpeg sox\n     ```\n   - **macOS** (需安装 Homebrew):\n     ```bash\n     brew install ffmpeg sox\n     ```\n   - **Windows**: 请下载并安装 [ffmpeg](https:\u002F\u002Fffmpeg.org\u002Fdownload.html) 和 [SoX](http:\u002F\u002Fsox.sourceforge.net\u002F)，并将它们添加到系统环境变量 PATH 中。\n\n3. **创建虚拟环境并安装 Python 依赖**\n   ```bash\n   python -m venv mir_env\n   source mir_env\u002Fbin\u002Factivate  # Windows 用户请使用: mir_env\\Scripts\\activate\n   \n   pip install -r requirements.txt\n   ```\n   *(注：若根目录无 `requirements.txt`，请安装核心库)*\n   ```bash\n   pip install numpy scipy librosa matplotlib mir_eval soundfile jupyter\n   ```\n\n4. **启动 Jupyter**\n   ```bash\n   jupyter notebook\n   ```\n   在浏览器中打开 `content` 目录下的笔记本文件。\n\n## 基本使用\n\n本项目的核心是通过执行 Notebook 中的代码块来学习 MIR 概念。以下是一个最简单的示例，演示如何加载音频并提取基本特征。\n\n### 示例：加载音频并绘制波形\n\n1. 打开 `content\u002F1_introduction\u002Fipython_audio.ipynb` 或新建一个 Notebook。\n2. 导入必要的库并加载音频文件：\n\n```python\nimport librosa\nimport matplotlib.pyplot as plt\nimport numpy as np\n\n# 加载示例音频 (librosa 自带示例)\ny, sr = librosa.load(librosa.ex('trumpet'))\n\n# 绘制波形图\nplt.figure(figsize=(10, 4))\nplt.plot(np.linspace(0, len(y)\u002Fsr, num=len(y)), y)\nplt.title('Audio Waveform')\nplt.xlabel('Time (s)')\nplt.ylabel('Amplitude')\nplt.tight_layout()\nplt.show()\n\n# 播放音频 (仅在 Jupyter 环境中有效)\nimport IPython.display as ipd\nipd.Audio(y, rate=sr)\n```\n\n3. 按 `Shift + Enter` 运行单元格，即可看到波形图并听到声音。\n\n### 学习路径建议\n- **入门**：从 `1_introduction` 章节开始，熟悉 Python 基础和 Jupyter 操作。\n- **核心概念**：进入 `2_music_representations` 和 `3_signal_analysis`，学习乐谱表示、傅里叶变换、色谱图（Chroma）等核心特征。\n- **进阶应用**：探索 `4_rhythm_tempo_beat`（节奏与节拍追踪）和 `5_machine_learning`（机器学习分类）。\n- **评估与同步**：参考 `6_evaluation` 学习如何使用 `mir_eval` 评估算法效果，或在 `7_music_synchronization` 中学习动态时间规整（DTW）。","某音乐科技公司的算法工程师正在开发一款能自动识别歌曲调性并生成可视化分析报告的 AI 应用。\n\n### 没有 musicinformationretrieval.com 时\n- 工程师需从零摸索音频信号处理理论，难以快速理解傅里叶变换、短时傅里叶变换（STFT）等核心数学概念在音乐中的具体含义。\n- 面对复杂的音频特征提取任务（如过零率、能量分析），缺乏标准的代码实现参考，导致反复调试底层库函数，开发效率极低。\n- 不清楚如何将 MIDI 音符准确转换为频率，或在不同律制间进行换算，容易在基础数据转换环节引入难以排查的误差。\n- 缺少对“色谱特征（Chroma）”等高级音乐表示法的直观演示，难以向非技术背景的产品经理解释算法原理。\n\n### 使用 musicinformationretrieval.com 后\n- 通过其交互式 Jupyter 笔记本，工程师可直接运行并修改关于傅里叶变换和频谱图的教程代码，瞬间掌握信号分析的核心逻辑。\n- 利用现成的特征提取示例（如能量、过零率计算），快速构建出稳定的数据处理流水线，将原本数天的探索工作缩短至几小时。\n- 直接调用项目中提供的 MIDI 与频率转换表及调律系统指南，确保了基础数据处理的准确性，避免了重复造轮子。\n- 借助“通过声音合成理解音频特征”等可视化章节，生动地展示了抽象特征的实际听感，轻松完成了跨部门的技术方案沟通。\n\nmusicinformationretrieval.com 将晦涩的音乐信息检索理论转化为可执行的代码实验，极大地降低了音频 AI 开发的门槛与试错成本。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fmusicinformationretrieval_musicinformationretrieval.com_ebca9dcb.png","musicinformationretrieval","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fmusicinformationretrieval_fa086a8c.png",null,"https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval",[78,82,86,89,92,95],{"name":79,"color":80,"percentage":81},"Jupyter Notebook","#DA5B0B",100,{"name":83,"color":84,"percentage":85},"Python","#3572A5",0,{"name":87,"color":88,"percentage":85},"Makefile","#427819",{"name":90,"color":91,"percentage":85},"CSS","#663399",{"name":93,"color":94,"percentage":85},"Shell","#89e051",{"name":96,"color":97,"percentage":85},"TeX","#3D6117",1271,414,"2026-04-05T02:45:13","MIT","Linux, macOS, Windows","未说明",{"notes":105,"python":103,"dependencies":106},"该项目主要作为基于 Jupyter Notebook 的教学资源，官方推荐通过 Binder 在云端一致的计算环境中直接运行，无需本地配置复杂环境。若需本地运行，需安装音频处理工具 SoX 和 ffmpeg，以及基础科学计算库（NumPy, SciPy）。内容涵盖音乐信息检索的基础信号分析、机器学习及评估方法，未提及对高性能 GPU 或大内存的特殊需求。",[107,108,109,110,111,112,113],"NumPy","SciPy","Jupyter","SoX","ffmpeg","mir_eval","ADTLib",[14,115,116],"其他","音频",[118,119,120,121,122,123],"ipython-notebook","music-information-retrieval","python","machine-learning","music","jupyter-notebook","2026-03-27T02:49:30.150509","2026-04-06T22:02:51.301201",[127,132,137,142,147,152],{"id":128,"question_zh":129,"answer_zh":130,"source_url":131},20155,"所有笔记本都兼容 Google Colab 了吗？","是的，所有笔记本已更新以兼容 Google Colab。这包括修复了依赖导入、资源文件加载路径以及可视化函数的调用方式（如将 waveplot 改为 waveshow）。用户现在可以直接在 Colab 中打开并运行这些笔记本，无需本地配置环境。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F47",{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},20150,"导入 stanford_mir 时出现 'No module named stanford_mir' 错误怎么办？","该库并未作为标准包发布，因此无法直接通过 pip 安装。解决方法是：\n1. 从原始链接下载 Python 脚本：https:\u002F\u002Fraw.githubusercontent.com\u002Fstevetjoa\u002Fmusicinformationretrieval.com\u002Fgh-pages\u002Fstanford_mir.py\n2. 将该文件保存为 `stanford_mir.py` 并放入你的工作目录或 Python 路径中。\n3. 或者，克隆整个 GitHub 仓库目录，进入该目录后直接从本地导入使用。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F53",{"id":138,"question_zh":139,"answer_zh":140,"source_url":141},20151,"如何在 Google Colab 中运行这些笔记本（Notebooks）？需要做什么修改？","为了在 Google Colab 中顺利运行，建议进行以下修改：\n1. 跳过 `stanford_mir` 的初始化（即使下载也会报错），注释掉相关代码：`# import stanford_mir; stanford_mir.init()`。\n2. 添加代码以自动下载音频素材文件，例如：`!wget -P \"audio\" \"https:\u002F\u002Fgithub.com\u002Fstevetjoa\u002Fmusicinformationretrieval.com\u002Fraw\u002Fgh-pages\u002Faudio\u002Fsimple_loop.wav\"`。\n3. 将已弃用的 `librosa.display.waveplot` 替换为新函数 `librosa.display.waveshow`。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F57",{"id":143,"question_zh":144,"answer_zh":145,"source_url":146},20152,"遇到 'AttributeError: _process_plot_var_args instance has no attribute prop_cycler' 错误如何解决？","此错误通常由 `librosa` 版本与 `matplotlib` 版本不兼容引起，特别是在 Python 2.7 环境下。解决方案包括：\n1. 升级到 Python 3 环境，该问题在 Python 3 中通常不存在。\n2. 将代码中的 `librosa.display.waveplot` 替换为 `librosa.display.waveshow`（新版 librosa 推荐用法）。\n3. 检查输入数据 `x` 是否格式正确，若问题依旧，可能是库版本冲突，建议更新 `librosa` 和 `matplotlib`。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F37",{"id":148,"question_zh":149,"answer_zh":150,"source_url":151},20153,"scikit-learn 升级后出现 'cross_validation' 模块找不到的错误怎么办？","从 scikit-learn 0.18 版本开始，`sklearn.cross_validation` 模块已被移至 `sklearn.model_selection`。\n请将代码中的导入语句和调用方式修改为：\n原代码：`from sklearn.cross_validation import cross_val_score`\n新代码：`from sklearn.model_selection import cross_val_score`\n或者直接调用：`sklearn.model_selection.cross_val_score()`。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F40",{"id":153,"question_zh":154,"answer_zh":155,"source_url":156},20154,"项目未来计划如何管理依赖和环境以确保长期可运行性？","为了解决 Google Colab 难以固定包版本和加载资源的问题，项目正考虑迁移到更稳定的方案：\n1. **Binder**: 允许通过 Docker 镜像固定 Python 环境和包版本，资源文件可直接包含在仓库中，无需每次下载。\n2. **JupyterBook**: 作为中间方案，提供美观的前端，并支持一键启动 Colab、Binder 或 JupyterLite。\n3. **JupyterLite**: 另一种选择，直接在浏览器中运行代码，无需云端服务器，但安装包较为复杂且受限于用户本地机器资源。","https:\u002F\u002Fgithub.com\u002Fmusicinformationretrieval\u002Fmusicinformationretrieval.com\u002Fissues\u002F72",[158],{"id":159,"version":160,"summary_zh":161,"released_at":162},118188,"v0.1.0","自2014年6月23日（星期一）起。","2014-06-23T08:32:39"]