[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-snakers4--silero-models":3,"tool-snakers4--silero-models":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},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。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},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",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},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。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":79,"owner_website":79,"owner_url":81,"languages":82,"stars":91,"forks":92,"last_commit_at":93,"license":94,"difficulty_score":95,"env_os":96,"env_gpu":96,"env_ram":96,"env_deps":97,"category_tags":102,"github_topics":103,"view_count":124,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":125,"updated_at":126,"faqs":127,"releases":158},521,"snakers4\u002Fsilero-models","silero-models","Silero Models: pre-trained text-to-speech models made embarrassingly simple","silero-models 是一套轻量级且高效的预训练文本转语音（TTS）模型库。它主要解决了语音合成技术在实际应用中部署门槛高、依赖环境复杂的问题。通过极简的 API 设计，silero-models 让开发者能够在一行代码内完成模型加载与音频生成，大幅降低了集成难度。\n\n这套模型非常适合希望快速实现语音功能的开发者、进行语音实验的研究人员，以及需要原型验证的产品设计师。其技术亮点在于卓越的推理速度与自然的语音质量，即使在普通 CPU 上也能保持高速运行。最新的 V5 版本进一步增强了功能，支持 SSML 标记语言以控制语调，并在俄语场景中实现了自动重音和同形异义词处理。此外，它还覆盖了英语、俄语及部分印度语言，并提供灵活的部署选项，如 PyTorch Hub 或直接 pip 安装。尽管遵循非商业许可协议，silero-models 凭借其易用性和高质量，依然是开源社区中备受推崇的语音合成解决方案。"," [![Mailing list : test](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEmail-gray.svg?style=for-the-badge&logo=gmail)](mailto:hello@silero.ai) [![Mailing list : test](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTelegram-blue.svg?style=for-the-badge&logo=telegram)](https:\u002F\u002Ft.me\u002Fsilero_speech) [![License: CC BY-NC 4.0](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-CC%20BY--NC%204.0-lightgrey.svg?style=for-the-badge)](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002FLICENSE)\n\n[![PyPI version](https:\u002F\u002Fbadge.fury.io\u002Fpy\u002Fsilero.svg)](https:\u002F\u002Fbadge.fury.io\u002Fpy\u002Fsilero)\n\n![header](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fsnakers4_silero-models_readme_322d557c5d07.png)\n\n- [Silero Models](#silero-models)\n - [Installation and Basics](#installation-and-basics)\n - [Text-To-Speech](#text-to-speech)\n - [Models and Speakers](#models-and-speakers)\n - [V5](#v5)\n - [V5 CIS Base Models](#v5-cis-base-models)\n - [V5 CIS Ext Models](#v5-cis-ext-models)\n - [V4](#v4)\n - [V3](#v3)\n - [Dependencies](#dependencies)\n - [PyTorch](#pytorch)\n - [Standalone Use](#standalone-use)\n - [SSML](#ssml)\n - [Cyrillic languages v4](#cyrillic-languages-v4)\n - [Indic languages v4](#indic-languages-v4)\n - [Example](#example)\n - [Supported languages](#supported-languages)\n - [Contact](#contact)\n - [Licence](#licence) \n - [Citations](#citations)\n - [Further reading](#further-reading)\n - [English](#english)\n - [Chinese](#chinese)\n - [Russian](#russian)\n\n# Silero Models\n\nOur TTS models satisfy the following criteria:\n\n- Fully end-to-end;\n- Large library of voices;\n- Natural-sounding speech;\n- One-line usage, minimal, portable;\n- Impressively fast on CPU and GPU;\n- For the Russian language - automated stress and homographs;\n \n## Installation and Basics\n\nYou can basically use our models in 3 flavours:\n\n- Via PyTorch Hub: `torch.hub.load()`;\n- Via pip: `pip install silero` and then `from silero import silero_tts`;\n- Via caching the required models and utils manually and modifying if necessary;\n\nModels are downloaded on demand both by pip and PyTorch Hub. If you need caching, do it manually or via invoking a necessary model once (it will be downloaded to a cache folder). Please see these [docs](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Fhub.html#loading-models-from-hub) for more information.\n\nPyTorch Hub and pip package are based on the same code. All of the `torch.hub.load` examples can be used with the pip package via this basic change:\n\n```python\nfrom silero import silero_tts\nmodel, example_text = silero_tts(language='ru',\n speaker='v5_ru')\naudio = model.apply_tts(text=example_text)\n```\n\n## Text-To-Speech\n\n### Models and Speakers\n\nAll of the provided models are listed in the [models.yml](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fmodels.yml) file. Any metadata and newer versions will be added there.\n\n#### V5\n\nV5 models support [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML). Also see Colab examples for main SSML tag usage.\n\nRussian-only models support automated stress and homographs. `v5_2_ru` cointains minor fixes and removes numpy and scipy dependencies.\n\n`v5_3_ru` cointains minor fixes. `v5_4_ru` also supports questions.\n\n| ID | Speakers | Auto-stress \u002F Homographs \u002F Questions | Language | SR | Colab |\n| ------- | --------------------------------------------- | ----------- | -------------- | ------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v5_4_ru` | `aidar`, `baya`, `kseniya`, `xenia` | ✅ \u002F ✅ \u002F ✅ | `ru` (Russian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_3_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (Russian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_2_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (Russian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (Russian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n\n#### V5 CIS Base Models\n\n- All of the below models support `8000`, `24000`, `48000` sampling rates and contain no auto-stress or homographs;\n- `v5_cis_base` models assume that proper stress should be added for each word for all languages, i.e. `к+ошка`;\n- `v5_cis_base_nostress` models assume that proper stress should be added for each word ONLY for slavic languages (i.e. `ru`, `bel`, `ukr`); \n- All of the below models are published under `MIT` [licence](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002FLICENSE_CIS);\n- V5 UTMOS and throughput [metrics](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002F%D0%9F%D1%83%D0%B1%D0%BB%D0%B8%D1%87%D0%BD%D0%B0%D1%8F-%D0%B4%D0%BE%D0%BA%D1%83%D0%BC%D0%B5%D0%BD%D1%82%D0%B0%D1%86%D0%B8%D1%8F-%D0%BF%D0%BE-%D1%81%D0%BA%D0%BE%D1%80%D0%BE%D1%81%D1%82%D0%B8-%D0%B8-%D0%BA%D0%B0%D1%87%D0%B5%D1%81%D1%82%D0%B2%D1%83-%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D1%8B);\n- V5 models support [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML). Also see Colab examples for main SSML tag usage;\n- Use [cases](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002F%D0%9D%D0%B0%D0%BF%D1%80%D0%B0%D0%B2%D0%BB%D0%B5%D0%BD%D0%B8%D1%8F-%D0%BF%D1%80%D0%B8%D0%BA%D0%BB%D0%B0%D0%B4%D0%BD%D0%BE%D0%B3%D0%BE-%D0%B8%D1%81%D0%BF%D0%BE%D0%BB%D1%8C%D0%B7%D0%BE%D0%B2%D0%B0%D0%BD%D0%B8%D1%8F-%D0%BC%D0%BE%D0%B4%D0%B5%D0%BB%D0%B5%D0%B9-%D1%81%D0%B8%D0%BD%D1%82%D0%B5%D0%B7%D0%B0) for the model;\n- Minimal system requirements: a PyTorch-compatible system, a modern processor with AVX2 instruction set for x86\u002F64 platform. \n\n| ID | Speakers | Language | Colab |\n| ------------------------------------- | -------------------------------------------- | -------------------- | -------------------- |\n| `v5_cis_base`, `v5_cis_base_nostress` | `aze_gamat` | `aze` (Azerbaijani) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `hye_zara` | `hye` (Armenian) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bak_aigul`, `bak_alfia`, `bak_alfia2` | `bak` (Bashkir) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bak_miyau`, `bak_ramilia` | `bak` (Bashkir) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bel_anatoliy`, `bel_dmitriy`, `bel_larisa` | `bel` (Belarus) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kat_vika` | `kat` (Georgian) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kbd_eduard` | `kbd` (Kab.-Cherkes) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kaz_zhadyra`, `kaz_zhazira` | `kaz` (Kazakh) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `xal_kejilgan`, `xal_kermen` | `xal` (Kalmyk) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kir_nurgul` | `kir` (Kyrgyz) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `mdf_oksana` | `mdf` (Moksha) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | all of these speakers, but with `ru_` prefix | `ru` (Russian) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `tgk_onaoy`, `tgk_safarhuja` | `tgk` (Tajik) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `tat_albina`, `tat_marat` | `tat` (Tatar) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `udm_bogdan` | `udm` (Udmurt) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `uzb_saida` | `uzb` (Uzbek) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `ukr_igor`, `ukr_roman` | `ukr` (Ukrainian) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kjh_karina`, `kjh_sibday` | `kjh` (Khakas) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `chv_ekaterina` | `chv` (Chuvash) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `erz_alexandr` | `erz` (Erzya) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `sah_zinaida` | `sah` (Yakut) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) \n\n\n\u003Cdetails>\n \u003Csummary>Supported alphabets\u003C\u002Fsummary>\n\n\n Please note that Georgian and Armenian are in fact internally supported via direct translation into cyrillic script inside of the package. Azerbaijani and Uzbek support both alphabets (Cyrillic and Latin).\n\n| ID | Название | Алфавит(ы) |\n|-----|---------------|--------------------------------------------|\n| aze | `aze` (Azerbaijani) | abcçdeәfgğhxıijkqlmnoöprsştuüvyz |\n| aze | `aze` (Azerbaijani) | абвгғдеәжзиыјкҝлмноөпрстуүфхһчҹш |\n| hye | `hye` (Armenian) | աբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆև |\n| bak | `bak` (Bashkir) | абвгдежзийклмнопрстуфхцчшщъыьэюяёғҙҡңҫүһәө |\n| bel | `bel` (Belarus) | абвгдежзйклмнопрстуфхцчшыьэюяёіў |\n| kat | `kat` (Georgian) | აბგდევზთიკლმნოპჟრსტუფქღყშჩცძწჭხჯჰ |\n| kbd | `kbd` (Kab.-Cherkes) | абвгдежзийклмнопрстуфхцчшщъыьэюяёӏ |\n| kaz | `kaz` (Kazakh) | абвгдежзийклмнопрстуфхцчшщыьэюяіғқңүұһәө |\n| xal | `xal` (Kalmyk) | абвгдежзийклмнопрстуфхцчшщъыьэюяҗңүһәө |\n| kir | `kir` (Kyrgyz) | абвгдежзийклмнопрстуфхцчшыьэюяёңүө |\n| mdf | `mdf` (Moksha) | абвгдежзийклмнопрстуфхцчшщъыьэюяё |\n| ru | `ru` (Russian) | абвгдеёжзийклмнопрстуфхцчшщъыьэюя |\n| tgk | `tgk` (Tajik) | абвгдежзийклмнопрстуфхчшъэюяёғқҳҷӣӯ |\n| tat | `tat` (Tatar) | абвгдежзийклмнопрстуфхцчшъыьэюяҗңүһәө |\n| udm | `udm` (Udmurt) | абвгдежзийклмнопрстуфхцчшщъыьэюяёӝӟӥӧӵ |\n| uzb | `uzb` (Uzbek) | абвгдежзийклмнопрстуфхцчшъьэюяёўғқҳ |\n| uzb | `uzb` (Uzbek) | abcdefghijklmnopqrstuvxyz |\n| ukr | `ukr` (Ukrainian) | абвгґдеєжзиіїйклмнопрстуфхцчшщьюя |\n| kjh | `kjh` (Khakas) | абвгдежзийклмнопрстуфхцчшщъыьэюяёіғңҷӧӱ |\n| chv | `chv` (Chuvash) | абвгдежзийклмнопрстуфхцчшщъыьэюяёҫӑӗӳ |\n| erz | `erz` (Erzya) | абвгдежзийклмнопрстуфхцчшщъыьэюяё |\n| sah | `sah` (Yakut) | абвгдежзийклмнопрстуфхцчшщъыьэюяёҕҥүһө |\n \n\u003C\u002Fdetails>\n\n#### V5 CIS Ext Models\n\n- All of the below models support `8000`, `24000`, `48000` sampling rates and contain no auto-stress or homographs;\n- `v5_cis_ext` models assume that proper stress should be added for each word for all languages, i.e. `к+ошка`;\n- `v5_cis_ext_nostress` are coming soon;\n- All of the below models are published under `CC-NC-BY` licence;\n- V5 models support [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML). Also see Colab examples for main SSML tag usage.\n\n| ID | Speakers | Language | Colab |\n| ------------ | --------------------------------------------------------------------- | ----------------- | -------------------- |\n| `v5_cis_ext` | `kaz_abai`, `kaz_aidana`, `kaz_aisha`, `kaz_bakir`, `kaz_danara` | `kaz` (Kazakh) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `xal_delghir`, `xal_erdni` | `xal` (Kalmyk) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_adiba`, `tat_alsou`, `tat_amir`, `tat_azat`, `tat_batir` | `tat` (Tatar) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_bulat`, `tat_damir`, `tat_guzel`, `tat_ildar`, `tat_ilgiz` | `tat` (Tatar) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_karim`, `tat_mansur`, `tat_murat`, `tat_rasima`, `tat_rustem` | `tat` (Tatar) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_timur`, `tat_zifa`, `tat_zufar`, `tat_zulfiya` | `tat` (Tatar) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `uzb_anora`, `uzb_dilnavoz` | `uzb` (Uzbek) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `ukr_kateryna`, `ukr_lada`, `ukr_mykyta`, `ukr_oleksa`, `ukr_tetiana` | `ukr` (Ukrainian) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `chv_aihwa`, `chv_alima` | `chv` (Chuvash) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n\n#### V4\n\nV4 models support [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML). Also see Colab examples for main SSML tag usage.\n\n\u003Cdetails>\n \u003Csummary>V4 models: v4_ru, v4_cyrillic, v4_ua, v4_uz, v4_indic \u003C\u002Fsummary>\n\n\n| ID | Speakers |Auto-stress | Language | SR | Colab |\n| ------------- | ----------- | ----------- |---------------------------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v4_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene`, `random` | yes | `ru` (Russian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v4_cyrillic`](#cyrillic-languages) | `b_ava`, `marat_tt`, `kalmyk_erdni`... | no | `cyrillic` [(Avar, Tatar, Kalmyk, ...)](#cyrillic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v4_ua` | `mykyta`, `random` | no | `ua` (Ukrainian) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v4_uz` | `dilnavoz` | no | `uz` (Uzbek) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v4_indic`](#indic-languages) | `hindi_male`, `hindi_female`, ..., `random` | no | `indic` [(Hindi, Telugu, ...)](#indic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n \n\u003C\u002Fdetails>\n\n#### V3\n\nV3 models support [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML). Also see Colab examples for main SSML tag usage.\n\n\u003Cdetails>\n \u003Csummary>V3 models: v3_en, v3_en_indic, v3_de, v3_es, v3_fr, v3_indic \u003C\u002Fsummary>\n\n\n| ID | Speakers |Auto-stress | Language | SR | Colab |\n| ------------- | ----------- | ----------- |---------------------------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v3_en` | `en_0`, `en_1`, ..., `en_117`, `random` | no | `en` (English) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_en_indic` | `tamil_female`, ..., `assamese_male`, `random` | no | `en` (English) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_de` | `eva_k`, ..., `karlsson`, `random` | no | `de` (German) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_es` | `es_0`, `es_1`, `es_2`, `random` | no | `es` (Spanish) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_fr` | `fr_0`, ..., `fr_5`, `random` | no | `fr` (French) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v3_indic`](#indic-languages) | `hindi_male`, `hindi_female`, ..., `random` | no | `indic` [(Hindi, Telugu, ...)](#indic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n\n\u003C\u002Fdetails>\n\n\n### Dependencies\n\nBasic dependencies for Colab examples:\n\n- `torch`, 1.10+ for v3 models\u002F 2.0+ for v4 and v5 models;\n- `torchaudio`, latest version bound to PyTorch should work (required only because models are hosted together with STT, not required for work);\n- `omegaconf`, latest (can be removed as well, if you do not load all of the configs);\n\n### PyTorch\n\n[![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb)\n\n[![Open on Torch Hub](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTorch-Hub-red?logo=pytorch&style=for-the-badge)](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fsnakers4_silero-models_tts\u002F)\n\n```python\n# V5\nimport torch\n\nlanguage = 'ru'\nmodel_id = 'v5_ru'\nsample_rate = 48000\nspeaker = 'xenia'\ndevice = torch.device('cpu')\n\nmodel, example_text = torch.hub.load(repo_or_dir='snakers4\u002Fsilero-models',\n model='silero_tts',\n language=language,\n speaker=model_id)\nmodel.to(device) # gpu or cpu\n\naudio = model.apply_tts(text=example_text,\n speaker=speaker,\n sample_rate=sample_rate)\n```\n\n### Standalone Use\n\n- Standalone usage only requires PyTorch 1.12+ and the Python Standard Library;\n- Please see the detailed examples in Colab;\n\n```python\n# V5\nimport os\nimport torch\n\ndevice = torch.device('cpu')\ntorch.set_num_threads(4)\nlocal_file = 'model.pt'\n\nif not os.path.isfile(local_file):\n torch.hub.download_url_to_file('https:\u002F\u002Fmodels.silero.ai\u002Fmodels\u002Ftts\u002Fru\u002Fv5_ru.pt',\n local_file) \n\nmodel = torch.package.PackageImporter(local_file).load_pickle(\"tts_models\", \"model\")\nmodel.to(device)\n\nexample_text = 'Меня зовут Лева Королев. Я из готов. И я уже готов открыть все ваши замки любой сложности!'\nsample_rate = 48000\nspeaker='baya'\n\naudio_paths = model.save_wav(text=example_text,\n speaker=speaker,\n sample_rate=sample_rate)\n```\n\n### SSML\n\nCheck out our [TTS Wiki page.](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)\n\n### Cyrillic languages v4\n\n> To be superseded with v5 model(s) soon.\n\nSupported tokenset:\n`!,-.:?iµöабвгдежзийклмнопрстуфхцчшщъыьэюяёђѓєіјњћќўѳғҕҗҙқҡңҥҫүұҳҷһӏӑӓӕӗәӝӟӥӧөӱӳӵӹ `\n\n| Speaker_ID | Language | Gender |\n| ------------ | --------------- | ------ |\n| b_ava | Avar | F |\n| b_bashkir | Bashkir | M |\n| b_bulb | Bulgarian | M |\n| b_bulc | Bulgarian | M |\n| b_che | Chechen | M |\n| b_cv | Chuvash | M |\n| cv_ekaterina | Chuvash | F |\n| b_myv | Erzya | M |\n| b_kalmyk | Kalmyk | M |\n| b_krc | Karachay-Balkar | M |\n| kz_M1 | Kazakh | M |\n| kz_M2 | Kazakh | M |\n| kz_F3 | Kazakh | F |\n| kz_F1 | Kazakh | F |\n| kz_F2 | Kazakh | F |\n| b_kjh | Khakas | F |\n| b_kpv | Komi-Ziryan | M |\n| b_lez | Lezghian | M |\n| b_mhr | Mari | F |\n| b_mrj | Mari High | M |\n| b_nog | Nogai | F |\n| b_oss | Ossetic | M |\n| b_ru | Russian | M |\n| b_tat | Tatar | M |\n| marat_tt | Tatar | M |\n| b_tyv | Tuvinian | M |\n| b_udm | Udmurt | M |\n| b_uzb | Uzbek | M |\n| b_sah | Yakut | M |\n| kalmyk_erdni | Kalmyk | M |\n| kalmyk_delghir | Kalmyk | F |\n\n### Indic languages v4\n\n#### Example\n\n(!!!) All input sentences should be romanized to ISO format using [`aksharamukha`](https:\u002F\u002Faksharamukha.appspot.com\u002Fpython). An example for `hindi`:\n\n```python\n# V3\nimport torch\nfrom aksharamukha import transliterate\n\n# Loading model\nmodel, example_text = torch.hub.load(repo_or_dir='snakers4\u002Fsilero-models',\n model='silero_tts',\n language='indic',\n speaker='v4_indic')\n\norig_text = \"प्रसिद्द कबीर अध्येता, पुरुषोत्तम अग्रवाल का यह शोध आलेख, उस रामानंद की खोज करता है\"\nroman_text = transliterate.process('Devanagari', 'ISO', orig_text)\nprint(roman_text)\n\naudio = model.apply_tts(roman_text,\n speaker='hindi_male')\n```\n\n#### Supported languages\n\n| Language | Speakers | Romanization function\n-- | -- | --\nhindi | `hindi_female`, `hindi_male` | `transliterate.process('Devanagari', 'ISO', orig_text)`\nmalayalam | `malayalam_female`, `malayalam_male` |`transliterate.process('Malayalam', 'ISO', orig_text)`\nmanipuri | `manipuri_female` |`transliterate.process('Bengali', 'ISO', orig_text)`\nbengali | `bengali_female`, `bengali_male` | `transliterate.process('Bengali', 'ISO', orig_text)`\nrajasthani | `rajasthani_female`, `rajasthani_female` | `transliterate.process('Devanagari', 'ISO', orig_text)`\ntamil | `tamil_female`, `tamil_male` |`transliterate.process('Tamil', 'ISO', orig_text, pre_options=['TamilTranscribe'])`\ntelugu | `telugu_female`, `telugu_male` | `transliterate.process('Telugu', 'ISO', orig_text)`\ngujarati | `gujarati_female`, `gujarati_male` | `transliterate.process('Gujarati', 'ISO', orig_text)`\nkannada | `kannada_female`, `kannada_male` |`transliterate.process('Kannada', 'ISO', orig_text)`\n\n## Contact\n\nTry our models, create an [issue](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002Fnew), join our [chat](https:\u002F\u002Ft.me\u002Fsilero_speech), [email](mailto:hello@silero.ai) us, and read the latest [news](https:\u002F\u002Ft.me\u002Fsilero_news).\n\n## Licence\n\nAll of the models are published under the main repo license (i.e. CC-NC-BY) except for the `base` cis-tts models, which are under MIT.\n\n## Citations\n\n```bibtex\n@misc{Silero Models,\n author = {Silero Team},\n title = {Silero Models: pre-trained text-to-speech models made embarrassingly simple},\n year = {2025},\n publisher = {GitHub},\n journal = {GitHub repository},\n howpublished = {\\url{https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models}},\n commit = {insert_some_commit_here},\n email = {hello@silero.ai}\n}\n```\n\n## Further reading\n\n### English\n\n- STT:\n - Towards an Imagenet Moment For Speech-To-Text - [link](https:\u002F\u002Fthegradient.pub\u002Ftowards-an-imagenet-moment-for-speech-to-text\u002F)\n - A Speech-To-Text Practitioners Criticisms of Industry and Academia - [link](https:\u002F\u002Fthegradient.pub\u002Fa-speech-to-text-practitioners-criticisms-of-industry-and-academia\u002F)\n - Modern Google-level STT Models Released - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F519562\u002F)\n\n- TTS:\n - Multilingual Text-to-Speech Models for Indic Languages - [link](https:\u002F\u002Fwww.analyticsvidhya.com\u002Fblog\u002F2022\u002F06\u002Fmultilingual-text-to-speech-models-for-indic-languages\u002F)\n - Our new public speech synthesis in super-high quality, 10x faster and more stable - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F660571\u002F)\n - High-Quality Text-to-Speech Made Accessible, Simple and Fast - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F549482\u002F)\n\n- VAD:\n - One Voice Detector to Rule Them All - [link](https:\u002F\u002Fthegradient.pub\u002Fone-voice-detector-to-rule-them-all\u002F)\n - Modern Portable Voice Activity Detector Released - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F537276\u002F)\n\n- Text Enhancement:\n - We have published a model for text repunctuation and recapitalization for four languages - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F581960\u002F)\n\n### Chinese\n\n- STT:\n - 迈向语音识别领域的 ImageNet 时刻 - [link](https:\u002F\u002Fwww.infoq.cn\u002Farticle\u002F4u58WcFCs0RdpoXev1E2)\n - 语音领域学术界和工业界的七宗罪 - [link](https:\u002F\u002Fwww.infoq.cn\u002Farticle\u002FlEe6GCRjF1CNToVITvNw)\n\n### Russian\n\n- STT\n - OpenAI решили распознавание речи! Разбираемся так ли это … - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F689572\u002F)\n - Наши сервисы для бесплатного распознавания речи стали лучше и удобнее - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F654227\u002F)\n - Telegram-бот Silero бесплатно переводит речь в текст - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F591563\u002F)\n - Бесплатное распознавание речи для всех желающих - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F587512\u002F)\n - Последние обновления моделей распознавания речи из Silero Models - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F577630\u002F)\n - Сжимаем трансформеры: простые, универсальные и прикладные способы cделать их компактными и быстрыми - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F563778\u002F)\n - Ультимативное сравнение систем распознавания речи: Ashmanov, Google, Sber, Silero, Tinkoff, Yandex - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F559640\u002F)\n - Мы опубликовали современные STT модели сравнимые по качеству с Google - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F519564\u002F)\n - Понижаем барьеры на вход в распознавание речи - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F494006\u002F)\n - Огромный открытый датасет русской речи версия 1.0 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F474462\u002F)\n - Насколько Быстрой Можно Сделать Систему STT? - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F531524\u002F)\n - Наша система Speech-To-Text - [link](https:\u002F\u002Fwww.silero.ai\u002Ftag\u002Four-speech-to-text\u002F)\n - Speech-To-Text - [link](https:\u002F\u002Fwww.silero.ai\u002Ftag\u002Fspeech-to-text\u002F)\n\n- TTS:\n - Делаем быстрый, качественный и доступный синтез на языках России — нужно ваше участие - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F872474\u002F)\n - Мы решили задачу омографов и ударений в русском языке - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F955130\u002F)\n - Теперь наш синтез также доступен в виде бота в Телеграме - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F682188\u002F)\n - Может ли синтез речи обмануть систему биометрической идентификации? - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F673996\u002F)\n - Теперь наш синтез на 20 языках - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F669910\u002F)\n - Теперь наш публичный синтез в супер-высоком качестве, в 10 раз быстрее и без детских болячек - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F660565\u002F)\n - Синтезируем голос бабушки, дедушки и Ленина + новости нашего публичного синтеза - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F584750\u002F)\n - Мы сделали наш публичный синтез речи еще лучше - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F563484\u002F)\n - Мы Опубликовали Качественный, Простой, Доступный и Быстрый Синтез Речи - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F549480\u002F)\n\n- VAD:\n - Новый релиз публичного детектора голоса Silero VAD v6 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F940750\u002F)\n - Наш публичный детектор голоса стал лучше - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F695738\u002F)\n - А ты используешь VAD? Что это такое и зачем он нужен - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F594745\u002F)\n - Модели для Детекции Речи, Чисел и Распознавания Языков - [link](https:\u002F\u002Fwww.silero.ai\u002Fvad-lang-classifier-number-detector\u002F)\n - Мы опубликовали современный Voice Activity Detector и не только -[link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F537274\u002F)\n\n- Text Enhancement:\n - Восстановление знаков пунктуации и заглавных букв — теперь и на длинных текстах - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F594565\u002F)\n - Мы опубликовали модель, расставляющую знаки препинания и заглавные буквы в тексте на четырех языках - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F581946\u002F)\n","[![邮件列表：测试](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEmail-gray.svg?style=for-the-badge&logo=gmail)](mailto:hello@silero.ai) [![邮件列表：测试](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTelegram-blue.svg?style=for-the-badge&logo=telegram)](https:\u002F\u002Ft.me\u002Fsilero_speech) [![许可证：CC BY-NC 4.0](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-CC%20BY--NC%204.0-lightgrey.svg?style=for-the-badge)](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002FLICENSE)\n\n[![PyPI 版本](https:\u002F\u002Fbadge.fury.io\u002Fpy\u002Fsilero.svg)](https:\u002F\u002Fbadge.fury.io\u002Fpy\u002Fsilero)\n\n![标题图](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fsnakers4_silero-models_readme_322d557c5d07.png)\n\n- [Silero 模型](#silero-models)\n - [安装与基础](#installation-and-basics)\n - [文本转语音](#text-to-speech)\n - [模型与说话人](#models-and-speakers)\n - [V5](#v5)\n - [V5 CIS 基础模型](#v5-cis-base-models)\n - [V5 CIS 扩展模型](#v5-cis-ext-models)\n - [V4](#v4)\n - [V3](#v3)\n - [依赖项](#dependencies)\n - [PyTorch Hub](#pytorch)\n - [独立使用](#standalone-use)\n - [SSML](#ssml)\n - [西里尔语 v4](#cyrillic-languages-v4)\n - [印地语系语言 v4](#indic-languages-v4)\n - [示例](#example)\n - [支持的语言](#supported-languages)\n - [联系方式](#contact)\n - [许可](#licence) \n - [引用](#citations)\n - [进一步阅读](#further-reading)\n - [英文](#english)\n - [中文](#chinese)\n - [俄文](#russian)\n\n# Silero 模型\n\n我们的 TTS（文本转语音）模型满足以下标准：\n\n- 完全端到端；\n- 丰富的声音库；\n- 听起来自然的语音；\n- 一行代码即可使用，简洁且便携；\n- 在 CPU 和 GPU 上速度惊人；\n- 针对俄语 - 自动重音和同形异义词处理；\n \n## 安装与基础\n\n您基本上可以通过以下 3 种方式使用我们的模型：\n\n- 通过 PyTorch Hub：`torch.hub.load()`；\n- 通过 pip：`pip install silero` 然后 `from silero import silero_tts`；\n- 通过手动缓存所需的模型和工具，并在必要时进行修改；\n\n通过 pip 和 PyTorch Hub 按需下载模型。如果您需要缓存，请手动操作或调用一次必要的模型（它将被下载到缓存文件夹）。有关更多信息，请参阅这些 [文档](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Fhub.html#loading-models-from-hub)。\n\nPyTorch Hub 和 pip 包基于相同的代码。所有的 `torch.hub.load` 示例都可以通过以下基本更改与 pip 包一起使用：\n\n```python\nfrom silero import silero_tts\nmodel, example_text = silero_tts(language='ru',\n speaker='v5_ru')\naudio = model.apply_tts(text=example_text)\n```\n\n## 文本转语音\n\n### 模型与说话人\n\n所有提供的模型均列在 [models.yml](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fmodels.yml) 文件中。任何元数据和新版本都将添加在那里。\n\n#### V5\n\nV5 模型支持 [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)（可扩展超文本标记语言）。另请参阅 Colab 示例以了解主要 SSML 标签的用法。\n\n仅俄语模型支持自动重音和同形异义词处理。`v5_2_ru` 包含 minor fixes 并移除了 numpy 和 scipy 依赖。\n\n`v5_3_ru` 包含 minor fixes。`v5_4_ru` 还支持疑问句。\n\n| ID | 说话人 | 自动重音 \u002F 同形异义词 \u002F 疑问句 | 语言 | 采样率 | Colab |\n| ------- | --------------------------------------------- | ----------- | -------------- | ------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v5_4_ru` | `aidar`, `baya`, `kseniya`, `xenia` | ✅ \u002F ✅ \u002F ✅ | `ru` (俄语) | `8000`, `24000`, `48000` | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_3_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (俄语) | `8000`, `24000`, `48000` | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_2_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (俄语) | `8000`, `24000`, `48000` | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v5_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene` | ✅ \u002F ✅ \u002F ❌ | `ru` (俄语) | `8000`, `24000`, `48000` | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n\n#### V5 CIS 基础模型\n\n- 以下所有模型均支持 `8000`, `24000`, `48000` 采样率，且不包含自动重音或同形异义词处理；\n- `v5_cis_base` 模型假设所有语言都需要为每个单词添加正确的重音，例如 `к+ошка`；\n- `v5_cis_base_nostress` 模型假设仅斯拉夫语言（即 `ru`, `bel`, `ukr`）需要为每个单词添加正确的重音； \n- 以下所有模型均在 `MIT` [许可](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002FLICENSE_CIS) 下发布；\n- V5 UTMOS 和吞吐量 [指标](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002F%D0%9F%D1%83%D0%B1%D0%BB%D0%B8%D1%87%D0%BD%D0%B0%D1%8F-%D0%B4%D0%BE%D0%BA%D1%83%D0%BC%D0%B5%D0%BD%D1%82%D0%B0%D1%86%D0%B8%D1%8F-%D0%BF%D0%BE-%D1%81%D0%BA%D0%BE%D1%80%D0%BE%D1%81%D1%82%D0%B8-%D0%B8-%D0%BA%D0%B0%D1%87%D0%B5%D1%81%D1%82%D0%B2%D1%83-%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D1%8B)；\n- V5 模型支持 [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)。另请参阅 Colab 示例以了解主要 SSML 标签的用法；\n- 使用 [指南](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002F%D0%9D%D0%B0%D0%BF%D1%80%D0%B0%D0%B2%D0%BB%D0%B5%D0%BD%D0%B8%D1%8F-%D0%BF%D1%80%D0%B8%D0%BA%D0%BB%D0%B0%D0%B4%D0%BD%D0%BE%D0%B3%D0%BE-%D0%B8%D1%81%D0%BF%D0%BE%D0%BB%D1%8C%D0%B7%D0%BE%D0%B2%D0%B0%D0%BD%D0%B8%D1%8F-%D0%BC%D0%BE%D0%B4%D0%B5%D0%BB%D0%B5%D0%B9-%D1%81%D0%B8%D0%BD%D1%82%D0%B5%D0%B7%D0%B0) 进行模型使用；\n- 最低系统要求：兼容 PyTorch 的系统，具有 AVX2 指令集的现代处理器用于 x86\u002F64 平台。\n\n| ID | 说话人 | 语言 | Colab |\n| ------------------------------------- | -------------------------------------------- | ------------------ | -------------------- |\n| `v5_cis_base`, `v5_cis_base_nostress` | `aze_gamat` | `aze` (阿塞拜疆语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `hye_zara` | `hye` (亚美尼亚语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bak_aigul`, `bak_alfia`, `bak_alfia2` | `bak` (巴什基尔语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bak_miyau`, `bak_ramilia` | `bak` (巴什基尔语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `bel_anatoliy`, `bel_dmitriy`, `bel_larisa` | `bel` (白俄罗斯语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kat_vika` | `kat` (格鲁吉亚语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kbd_eduard` | `kbd` (卡巴尔达 - 切尔克斯语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kaz_zhadyra`, `kaz_zhazira` | `kaz` (哈萨克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `xal_kejilgan`, `xal_kermen` | `xal` (卡尔梅克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kir_nurgul` | `kir` (吉尔吉斯语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `mdf_oksana` | `mdf` (莫克沙语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | 上述所有说话人，但带有 `ru_` 前缀 | `ru` (俄语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `tgk_onaoy`, `tgk_safarhuja` | `tgk` (塔吉克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `tat_albina`, `tat_marat` | `tat` (鞑靼语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `udm_bogdan` | `udm` (乌德穆尔特语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `uzb_saida` | `uzb` (乌兹别克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `ukr_igor`, `ukr_roman` | `ukr` (乌克兰语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `kjh_karina`, `kjh_sibday` | `kjh` (哈卡斯语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `chv_ekaterina` | `chv` (楚瓦什语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `erz_alexandr` | `erz` (埃尔齐亚语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_base`, `v5_cis_base_nostress` | `sah_zinaida` | `sah` (雅库特语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n\n\n\u003Cdetails>\n \u003Csummary>支持的字母表\u003C\u002Fsummary>\n\n\n 请注意，格鲁吉亚语和亚美尼亚语实际上是通过在包内部直接转换为西里尔字母表来支持的。阿塞拜疆语和乌兹别克语支持两种字母表（西里尔字母和拉丁字母）。\n\n| ID | 名称 | 字母表 |\n|-----|---------------|--------------------------------------------|\n| aze | `aze` (阿塞拜疆语) | abcçdeәfgğhxıijkqlmnoöprsştuüvyz |\n| aze | `aze` (阿塞拜疆语) | абвгғдеәжзиыјкҝлмноөпрстуүфхһчҹш |\n| hye | `hye` (亚美尼亚语) | աբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆև |\n| bak | `bak` (巴什基尔语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёғҙҡңҫүһәө |\n| bel | `bel` (白俄罗斯语) | абвгдежзйклмнопрстуфхцчшыьэюяёіў |\n| kat | `kat` (格鲁吉亚语) | აბგდევზთიკლმნოპჟრსტუფქღყშჩცძწჭხჯჰ |\n| kbd | `kbd` (卡巴尔达 - 切尔克斯语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёӏ |\n| kaz | `kaz` (哈萨克语) | абвгдежзийклмнопрстуфхцчшщыьэюяіғқңүұһәө |\n| xal | `xal` (卡尔梅克语) | абвгдежзийклмнопрстуфхцчшщъыьэюяҗңүһәө |\n| kir | `kir` (吉尔吉斯语) | абвгдежзийклмнопрстуфхцчшыьэюяёңүө |\n| mdf | `mdf` (莫克沙语) | абвгдежзийклмнопрстуфхцчшщъыьэюяё |\n| ru | `ru` (俄语) | абвгдеёжзийклмнопрстуфхцчшщъыьэюя |\n| tgk | `tgk` (塔吉克语) | абвгдежзийклмнопрстуфхцчшъэюяёғқҳҷӣӯ |\n| tat | `tat` (鞑靼语) | абвгдежзийклмнопрстуфхцчшъыьэюяҗңүһәө |\n| udm | `udm` (乌德穆尔特语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёӝӟӥӧӵ |\n| uzb | `uzb` (乌兹别克语) | абвгдежзийклмнопрстуфхцчшъьэюяёўғқҳ |\n| uzb | `uzb` (乌兹别克语) | abcdefghijklmnopqrstuvxyz |\n| ukr | `ukr` (乌克兰语) | абвгґдеєжзиіїйклмнопрстуфхцчшщьюя |\n| kjh | `kjh` (哈卡斯语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёіғңҷӧӱ |\n| chv | `chv` (楚瓦什语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёҫӑӗӳ |\n| erz | `erz` (埃尔齐亚语) | абвгдежзийклмнопрстуфхцчшщъыьэюяё |\n| sah | `sah` (雅库特语) | абвгдежзийклмнопрстуфхцчшщъыьэюяёҕҥүһө |\n \n\u003C\u002Fdetails>\n\n#### V5 CIS Ext 模型\n\n- 以下所有模型均支持 `8000`、`24000`、`48000` 采样率，且不包含自动重音或同形异义词处理；\n- `v5_cis_ext` 模型假设所有语言的每个单词都应添加正确的重音，例如 `к+ошка`；\n- `v5_cis_ext_nostress` 即将推出；\n- 以下所有模型均基于 `CC-NC-BY` 许可证发布；\n- V5 模型支持 [SSML（语音合成标记语言）](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)。另请参阅 Colab 示例以了解主要 SSML 标签的用法。\n\n| ID | 说话人 | 语言 | Colab |\n| ------------ | --------------------------------------------------------------------- | ----------------- | -------------------- |\n| `v5_cis_ext` | `kaz_abai`, `kaz_aidana`, `kaz_aisha`, `kaz_bakir`, `kaz_danara` | `kaz` (哈萨克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `xal_delghir`, `xal_erdni` | `xal` (卡尔梅克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_adiba`, `tat_alsou`, `tat_amir`, `tat_azat`, `tat_batir` | `tat` (鞑靼语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_bulat`, `tat_damir`, `tat_guzel`, `tat_ildar`, `tat_ilgiz` | `tat` (鞑靼语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_karim`, `tat_mansur`, `tat_murat`, `tat_rasima`, `tat_rustem` | `tat` (鞑靼语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `tat_timur`, `tat_zifa`, `tat_zufar`, `tat_zulfiya` | `tat` (鞑靼语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `uzb_anora`, `uzb_dilnavoz` | `uzb` (乌兹别克语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `ukr_kateryna`, `ukr_lada`, `ukr_mykyta`, `ukr_oleksa`, `ukr_tetiana` | `ukr` (乌克兰语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n| `v5_cis_ext` | `chv_aihwa`, `chv_alima` | `chv` (楚瓦什语) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts_cis.ipynb) |\n\n#### V4\n\nV4 模型支持 [SSML](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)。另请参阅 Colab 示例以了解主要 SSML 标签的用法。\n\n\u003Cdetails>\n \u003Csummary>V4 模型：v4_ru, v4_cyrillic, v4_ua, v4_uz, v4_indic \u003C\u002Fsummary>\n\n| ID | 说话人 | 自动重音 | 语言 | 采样率 (SR) | Colab |\n| ------------- | ----------- | ----------- |---------------------------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v4_ru` | `aidar`, `baya`, `kseniya`, `xenia`, `eugene`, `random` | 是 | `ru` (俄语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v4_cyrillic`](#cyrillic-languages) | `b_ava`, `marat_tt`, `kalmyk_erdni`... | 否 | `cyrillic` [(阿瓦尔语，鞑靼语，卡尔梅克语，...)](#cyrillic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v4_ua` | `mykyta`, `random` | 否 | `ua` (乌克兰语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v4_uz` | `dilnavoz` | 否 | `uz` (乌兹别克语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v4_indic`](#indic-languages) | `hindi_male`, `hindi_female`, ..., `random` | 否 | `indic` [(印地语，泰卢固语，...)](#indic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n \n\u003C\u002Fdetails>\n\n#### V3\n\nV3 模型支持 [SSML (语音合成标记语言)](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)。此外，请查看 Colab 示例以了解主要 SSML 标签的用法。\n\n\u003Cdetails>\n \u003Csummary>V3 模型：v3_en, v3_en_indic, v3_de, v3_es, v3_fr, v3_indic \u003C\u002Fsummary>\n\n\n| ID | 说话人 | 自动重音 | 语言 | 采样率 (SR) | Colab |\n| ------------- | ----------- | ----------- |---------------------------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| `v3_en` | `en_0`, `en_1`, ..., `en_117`, `random` | 否 | `en` (英语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_en_indic` | `tamil_female`, ..., `assamese_male`, `random` | 否 | `en` (英语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_de` | `eva_k`, ..., `karlsson`, `random` | 否 | `de` (德语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_es` | `es_0`, `es_1`, `es_2`, `random` | 否 | `es` (西班牙语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| `v3_fr` | `fr_0`, ..., `fr_5`, `random` | 否 | `fr` (法语) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n| [`v3_indic`](#indic-languages) | `hindi_male`, `hindi_female`, ..., `random` | 否 | `indic` [(印地语，泰卢固语，...)](#indic-languages) | `8000`, `24000`, `48000` | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb) |\n\n\u003C\u002Fdetails>\n\n\n\n\n### 依赖项\n\nColab 示例的基本依赖项：\n\n- `torch`，v3 模型需要 1.10+ \u002F v4 和 v5 模型需要 2.0+；\n- `torchaudio`，绑定到 PyTorch 的最新版本即可（仅因为模型与 STT (语音识别) 托管在一起而需要，实际工作不需要）；\n- `omegaconf`，最新版本（如果不加载所有配置，也可以移除）；\n\n### PyTorch\n\n[![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples_tts.ipynb)\n\n[![Open on Torch Hub](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTorch-Hub-red?logo=pytorch&style=for-the-badge)](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fsnakers4_silero-models_tts\u002F)\n\n```python\n# V5\nimport torch\n\nlanguage = 'ru'\nmodel_id = 'v5_ru'\nsample_rate = 48000\nspeaker = 'xenia'\ndevice = torch.device('cpu')\n\nmodel, example_text = torch.hub.load(repo_or_dir='snakers4\u002Fsilero-models',\n model='silero_tts',\n language=language,\n speaker=model_id)\nmodel.to(device) # gpu or cpu\n\naudio = model.apply_tts(text=example_text,\n speaker=speaker,\n sample_rate=sample_rate)\n```\n\n### 独立使用\n\n- 独立使用仅需 PyTorch 1.12+ 和 Python 标准库；\n- 请参阅 Colab 中的详细示例；\n\n```python\n# V5\nimport os\nimport torch\n\ndevice = torch.device('cpu')\ntorch.set_num_threads(4)\nlocal_file = 'model.pt'\n\nif not os.path.isfile(local_file):\n torch.hub.download_url_to_file('https:\u002F\u002Fmodels.silero.ai\u002Fmodels\u002Ftts\u002Fru\u002Fv5_ru.pt',\n local_file) \n\nmodel = torch.package.PackageImporter(local_file).load_pickle(\"tts_models\", \"model\")\nmodel.to(device)\n\nexample_text = 'Меня зовут Лева Королев. Я из готов. И я уже готов открыть все ваши замки любой сложности!'\nsample_rate = 48000\nspeaker='baya'\n\naudio_paths = model.save_wav(text=example_text,\n speaker=speaker,\n sample_rate=sample_rate)\n```\n\n### SSML\n\n请查看我们的 [TTS (Text-to-Speech，文本转语音) Wiki 页面。](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FSSML)\n\n### 西里尔语言 v4\n\n> 即将被 v5 模型取代。\n\n支持的词元集：\n`!,-.:?iµöабвгдежзийклмнопрстуфхцчшщъыьэюяёђѓєіјњћќўѳғҕҗҙқҡңҥҫүұҳҷһӏӑӓӕӗәӝӟӥӧөӱӳӵӹ `\n\n| Speaker_ID | Language | Gender |\n| ------------ | --------------- | ------ |\n| b_ava | Avar | F |\n| b_bashkir | Bashkir | M |\n| b_bulb | Bulgarian | M |\n| b_bulc | Bulgarian | M |\n| b_che | Chechen | M |\n| b_cv | Chuvash | M |\n| cv_ekaterina | Chuvash | F |\n| b_myv | Erzya | M |\n| b_kalmyk | Kalmyk | M |\n| b_krc | Karachay-Balkar | M |\n| kz_M1 | Kazakh | M |\n| kz_M2 | Kazakh | M |\n| kz_F3 | Kazakh | F |\n| kz_F1 | Kazakh | F |\n| kz_F2 | Kazakh | F |\n| b_kjh | Khakas | F |\n| b_kpv | Komi-Ziryan | M |\n| b_lez | Lezghian | M |\n| b_mhr | Mari | F |\n| b_mrj | Mari High | M |\n| b_nog | Nogai | F |\n| b_oss | Ossetic | M |\n| b_ru | Russian | M |\n| b_tat | Tatar | M |\n| marat_tt | Tatar | M |\n| b_tyv | Tuvinian | M |\n| b_udm | Udmurt | M |\n| b_uzb | Uzbek | M |\n| b_sah | Yakut | M |\n| kalmyk_erdni | Kalmyk | M |\n| kalmyk_delghir | Kalmyk | F |\n\n### 印度语言 v4\n\n#### 示例\n\n(!!!) 所有输入句子都应使用 [`aksharamukha`](https:\u002F\u002Faksharamukha.appspot.com\u002Fpython) 罗马化为 ISO 格式。以 `hindi` 为例：\n\n```python\n# V3\nimport torch\nfrom aksharamukha import transliterate\n\n# Loading model\nmodel, example_text = torch.hub.load(repo_or_dir='snakers4\u002Fsilero-models',\n model='silero_tts',\n language='indic',\n speaker='v4_indic')\n\norig_text = \"प्रसिद्द कबीर अध्येता, पुरुषोत्तम अग्रवाल का यह शोध आलेख, उस रामानंद की खोज करता है\"\nroman_text = transliterate.process('Devanagari', 'ISO', orig_text)\nprint(roman_text)\n\naudio = model.apply_tts(roman_text,\n speaker='hindi_male')\n```\n\n#### 支持的语言\n\n| Language | Speakers | Romanization function\n-- | -- | --\nhindi | `hindi_female`, `hindi_male` | `transliterate.process('Devanagari', 'ISO', orig_text)`\nmalayalam | `malayalam_female`, `malayalam_male` |`transliterate.process('Malayalam', 'ISO', orig_text)`\nmanipuri | `manipuri_female` |`transliterate.process('Bengali', 'ISO', orig_text)`\nbengali | `bengali_female`, `bengali_male` | `transliterate.process('Bengali', 'ISO', orig_text)`\nrajasthani | `rajasthani_female`, `rajasthani_female` | `transliterate.process('Devanagari', 'ISO', orig_text)`\ntamil | `tamil_female`, `tamil_male` |`transliterate.process('Tamil', 'ISO', orig_text, pre_options=['TamilTranscribe'])`\ntelugu | `telugu_female`, `telugu_male` | `transliterate.process('Telugu', 'ISO', orig_text)`\ngujarati | `gujarati_female`, `gujarati_male` | `transliterate.process('Gujarati', 'ISO', orig_text)`\nkannada | `kannada_female`, `kannada_male` |`transliterate.process('Kannada', 'ISO', orig_text)`\n\n## 联系方式\n\n尝试使用我们的模型，创建一个 [issue](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002Fnew)，加入我们的 [chat](https:\u002F\u002Ft.me\u002Fsilero_speech)，[email](mailto:hello@silero.ai) 联系我们，并阅读最新的 [news](https:\u002F\u002Ft.me\u002Fsilero_news)。\n\n## 许可证\n\n除 `base` cis-tts 模型（采用 MIT 许可）外，所有模型均根据主仓库许可证发布（即 CC-NC-BY）。\n\n## 引用\n\n```bibtex\n@misc{Silero Models,\n author = {Silero Team},\n title = {Silero Models: pre-trained text-to-speech models made embarrassingly simple},\n year = {2025},\n publisher = {GitHub},\n journal = {GitHub repository},\n howpublished = {\\url{https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models}},\n commit = {insert_some_commit_here},\n email = {hello@silero.ai}\n}\n```\n\n## 延伸阅读\n\n### 英文\n\n- STT (Speech-To-Text，语音转文本):\n - Towards an Imagenet Moment For Speech-To-Text - [link](https:\u002F\u002Fthegradient.pub\u002Ftowards-an-imagenet-moment-for-speech-to-text\u002F)\n - A Speech-To-Text Practitioners Criticisms of Industry and Academia - [link](https:\u002F\u002Fthegradient.pub\u002Fa-speech-to-text-practitioners-criticisms-of-industry-and-academia\u002F)\n - Modern Google-level STT Models Released - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F519562\u002F)\n\n- TTS:\n - Multilingual Text-to-Speech Models for Indic Languages - [link](https:\u002F\u002Fwww.analyticsvidhya.com\u002Fblog\u002F2022\u002F06\u002Fmultilingual-text-to-speech-models-for-indic-languages\u002F)\n - Our new public speech synthesis in super-high quality, 10x faster and more stable - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F660571\u002F)\n - High-Quality Text-to-Speech Made Accessible, Simple and Fast - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F549482\u002F)\n\n- VAD (Voice Activity Detector，语音活动检测):\n - One Voice Detector to Rule Them All - [link](https:\u002F\u002Fthegradient.pub\u002Fone-voice-detector-to-rule-them-all\u002F)\n - Modern Portable Voice Activity Detector Released - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F537276\u002F)\n\n- 文本增强:\n - We have published a model for text repunctuation and recapitalization for four languages - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F581960\u002F)\n\n### 中文\n\n- STT:\n - 迈向语音识别领域的 ImageNet 时刻 - [link](https:\u002F\u002Fwww.infoq.cn\u002Farticle\u002F4u58WcFCs0RdpoXev1E2)\n - 语音领域学术界和工业界的七宗罪 - [link](https:\u002F\u002Fwww.infoq.cn\u002Farticle\u002FlEe6GCRjF1CNToVITvNw)\n\n### 俄语\n\n- STT (Speech-To-Text，语音转文本):\n - OpenAI 解决了语音识别问题！我们来分析一下是否真的如此 … - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F689572\u002F)\n - 我们的免费语音识别服务变得更优质、更便捷了 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F654227\u002F)\n - Telegram 机器人 Silero 免费将语音转换为文本 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F591563\u002F)\n - 面向所有希望者的免费语音识别 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F587512\u002F)\n - Silero Models 中最新的语音识别模型更新 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F577630\u002F)\n - 压缩 Transformer（一种深度学习模型）：使其紧凑快速的简单、通用及实用方法 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F563778\u002F)\n - 语音识别系统的终极对比：Ashmanov, Google, Sber, Silero, Tinkoff, Yandex - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F559640\u002F)\n - 我们发布了质量可与 Google 相媲美的现代 STT 模型 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F519564\u002F)\n - 降低语音识别的入门门槛 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F494006\u002F)\n - 巨大的俄语语音开源数据集 1.0 版 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F474462\u002F)\n - STT 系统能有多快？ - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F531524\u002F)\n - 我们的 Speech-To-Text（语音转文本）系统 - [link](https:\u002F\u002Fwww.silero.ai\u002Ftag\u002Four-speech-to-text\u002F)\n - Speech-To-Text（语音转文本） - [link](https:\u002F\u002Fwww.silero.ai\u002Ftag\u002Fspeech-to-text\u002F)\n\n- TTS (Text-To-Speech，文本转语音):\n - 我们正在为俄罗斯语言制作快速、高质量且易用的合成语音——需要您的参与 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F872474\u002F)\n - 我们解决了俄语中的同形异义词和重音问题 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F955130\u002F)\n - 现在我们的合成语音也可以通过 Telegram 机器人获取 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F682188\u002F)\n - 语音合成能否欺骗生物识别系统？ - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F673996\u002F)\n - 现在我们的合成支持 20 种语言 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F669910\u002F)\n - 现在我们的公开合成具有超高质量，速度快 10 倍，且没有早期版本的缺陷 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F660565\u002F)\n - 合成奶奶、爷爷和列宁的声音 + 我们公开合成的新闻 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F584750\u002F)\n - 我们将公开语音合成做得更好了 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F563484\u002F)\n - 我们发布了高质量、简单、易用且快速的语音合成 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F549480\u002F)\n\n- VAD (Voice Activity Detection，语音活动检测):\n - 公开语音检测器 Silero VAD v6 新版本发布 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F940750\u002F)\n - 我们的公开语音检测器变得更好了 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F695738\u002F)\n - 你使用 VAD 吗？它是什么以及为什么需要它 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F594745\u002F)\n - 用于语音检测、数字检测和语言识别的模型 - [link](https:\u002F\u002Fwww.silero.ai\u002Fvad-lang-classifier-number-detector\u002F)\n - 我们发布了现代的 Voice Activity Detector（语音活动检测器）以及其他内容 -[link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F537274\u002F)\n\n- 文本增强:\n - 恢复标点符号和大写字母——现在也适用于长文本 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F594565\u002F)\n - 我们发布了一种模型，可在四种语言的文本中添加标点和大写 - [link](https:\u002F\u002Fhabr.com\u002Fru\u002Fpost\u002F581946\u002F)","# Silero Models 快速上手指南\n\n## 1. 环境准备\n- **系统要求**: 支持 PyTorch 的操作系统。\n- **硬件建议**: 现代处理器（x86\u002F64 平台需支持 AVX2 指令集以获得最佳性能），支持 CPU 和 GPU 推理。\n- **前置依赖**: Python 环境，PyTorch。\n\n## 2. 安装步骤\n提供两种使用方式，推荐使用 pip 包管理：\n\n- **方式一：通过 pip 安装**\n ```bash\n pip install silero\n ```\n- **方式二：PyTorch Hub (无需安装)**\n 直接使用 `torch.hub.load()` 加载模型，模型会在首次使用时自动下载。\n\n## 3. 基本使用\n以下示例展示了如何使用 pip 安装的包进行文本转语音 (TTS) 生成。\n\n```python\nfrom silero import silero_tts\nmodel, example_text = silero_tts(language='ru',\n speaker='v5_ru')\naudio = model.apply_tts(text=example_text)\n```\n\n**说明：**\n- 支持多种语言，包括俄语、乌克兰语、哈萨克语等 CIS 地区语言及英语。\n- 模型文件会在首次调用时自动下载至缓存文件夹。\n- 详细模型列表及更多示例可参考官方文档。","某创业团队正在开发一款支持俄语和英语的新闻资讯朗读 App，急需在有限预算下快速集成高质量语音合成功能。\n\n### 没有 silero-models 时\n- 需要从零训练模型或购买昂贵的商业 API，资金压力大且上线周期长达数周。\n- 现有开源方案依赖复杂，部署时需要配置大量环境，CPU 推理速度慢导致用户等待感强。\n- 俄语发音缺乏自动重音处理，合成语音听起来生硬机械，严重影响用户体验。\n- 代码集成繁琐，难以适配移动端或边缘设备，后续维护工作量巨大。\n\n### 使用 silero-models 后\n- silero-models 提供预训练模型，通过一行代码即可调用，将开发周期缩短至几天。\n- 推理速度极快，在普通 CPU 上也能实现实时语音生成，无需依赖昂贵的 GPU 硬件。\n- 内置俄语自动重音和同形异义词处理，显著提升了多语言播报的自然度和准确度。\n- 支持多种轻量级安装方式，方便打包进应用，极大降低了后续运维和更新成本。\n\nsilero-models 让开发者能以最低成本获得高性能、多语言的语音合成能力，真正实现端到端的快速落地。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fsnakers4_silero-models_24bdc1ac.png","snakers4","Alexander Veysov","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fsnakers4_2d4de121.png","It is by will alone I set my mind in motion.",null,"aveysov@gmail.com","https:\u002F\u002Fgithub.com\u002Fsnakers4",[83,87],{"name":84,"color":85,"percentage":86},"Jupyter Notebook","#DA5B0B",98.3,{"name":88,"color":89,"percentage":90},"Python","#3572A5",1.7,5853,362,"2026-04-04T15:28:53","NOASSERTION",1,"未说明",{"notes":98,"python":96,"dependencies":99},"1. 最小系统要求为 PyTorch 兼容系统，x86\u002F64 平台需现代处理器支持 AVX2 指令集。2. 模型按需下载，首次运行后自动缓存。3. 新版（如 v5_2_ru）已移除 numpy 和 scipy 依赖。4. 支持 SSML 标签控制语音合成。5. 在 CPU 和 GPU 上均有良好性能表现。6. 包含俄语及独联体国家语言模型。",[100,101],"torch","silero",[55,13],[104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123],"speech-to-text","pretrained-models","pytorch","colab","torch-hub","text-to-speech","tts-models","speech","speech-synthesis","tts","kazakh","russian","ukrainian","uzbek","azerbaijani","belarus","tajik","kyrgyz","armenian","georgian",9,"2026-03-27T02:49:30.150509","2026-04-06T07:12:53.977116",[128,133,138,143,148,153],{"id":129,"question_zh":130,"answer_zh":131,"source_url":132},2094,"独立使用（Standalone）是否必须安装 NumPy？","文档说明仅需 PyTorch 1.10+ 和 Python 标准库，但实际运行可能需要 NumPy。维护者表示将在下个版本修复文档。如果希望移除 NumPy 依赖，可以参考社区方案，使用 Python 内置的 `array` 模块替代（例如在 TTS server 脚本中处理）。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F163",{"id":134,"question_zh":135,"answer_zh":136,"source_url":137},2095,"遇到 \"RuntimeError: Unknown qengine\" 错误如何解决？如何获取本地模型文件？","建议尝试新版本 `v3_1_ru`，其中量化方式有所改变。若需手动获取本地模型，可查看 `models.yml` 文件中的 URL 并自行下载。如果 `torch.hub.download_url_to_file` 持续报错，可尝试直接从 `silero` 包导入模型。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F148",{"id":139,"question_zh":140,"answer_zh":141,"source_url":142},2096,"Windows 上通过 torch.hub 加载 STT 模型失败怎么办？","这通常与缓存路径有关。请在 Python 解释器中检查以下路径是否存在：`C:\\Users\\...\\.cache\\torch\\hub\\snakers4_silero-models_master\\model\\en_v5.jit`。运行 `torch.hub.get_dir()` 确认缓存目录。`torch.hub` 仅为便利加载，文件也可手动下载。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F121",{"id":144,"question_zh":145,"answer_zh":146,"source_url":147},2097,"如何获取预训练模型的中间层输出？","这些模型并非设计为可随意修改的。如果确实需要修改以获取特定层（如 FC 层之前）的输出，需要手动查看容器内部结构（它们看起来像文件夹），并在 forward pass 过程中修改代码以返回指定点的输出。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F51",{"id":149,"question_zh":150,"answer_zh":151,"source_url":152},2098,"TensorFlow 模型导出为单个 protobuf 文件需要额外步骤吗？","不需要，只需调用一个函数即可导出为单个 protobuf 文件。如果导出异常，可能是 `onnx-tensorflow` 相关的问题（例如 PR #603 合并后的影响），建议检查相关依赖版本。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F5",{"id":154,"question_zh":155,"answer_zh":156,"source_url":157},2099,"能否通过 SSML 格式调整句子间的语调衰减和停顿？","Pitch 和速度参数可以通过 SSML 手动控制，但直接设置可能导致伪影。当前系统已针对较好声音微调了暂停时的音调插值。若要彻底解决多句合成时的语调问题，最可靠的方法是使用包含正确说话节奏的新数据集重新训练模型。","https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fissues\u002F164",[159,163,167,172,176,181,186,191,196,201],{"id":160,"version":161,"summary_zh":79,"released_at":162},101535,"v5.5","2026-02-03T16:53:21",{"id":164,"version":165,"summary_zh":79,"released_at":166},101536,"v5.4","2026-01-30T09:18:51",{"id":168,"version":169,"summary_zh":170,"released_at":171},101537,"v5.2","- Added support for 19 new languages from Russia and the CIS;\r\n- Added `v5_cis_base`, `v5_cis_base_nostress`, `v5_cis_ext` models;\r\n\r\n## What's Changed\r\n\r\n* Fx by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F301\r\n* Rm v2 and v1 by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F302\r\n* TTS v5 CIS by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F306\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fcompare\u002Fv5.1...v5.2","2025-11-22T18:35:50",{"id":173,"version":174,"summary_zh":79,"released_at":175},101538,"v5.1","2025-10-30T17:08:22",{"id":177,"version":178,"summary_zh":179,"released_at":180},101539,"v5.0","- Legacy tools and models deprecated;\r\n- `v5_ru` TTS models released with homographs, higher quality and improved speed;\r\n- Preparations for massive CIS TTS model release;","2025-10-30T16:54:22",{"id":182,"version":183,"summary_zh":184,"released_at":185},101540,"v0.4.1","## What's Changed\r\n* Fix models.yml loading by @rominf in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F162\r\n\r\n## New Contributors\r\n* @rominf made their first contribution in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F162\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fcompare\u002Fv0.4...v0.4.1","2022-06-12T18:34:51",{"id":187,"version":188,"summary_zh":189,"released_at":190},101541,"v0.4","## What's Changed\r\n\r\n* Add version 3.1 by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F157\r\n* Fx by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F158\r\n* Fx by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F159\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fcompare\u002Fv0.3...v0.4","2022-06-06T18:32:29",{"id":192,"version":193,"summary_zh":194,"released_at":195},101542,"v0.3","## What's Changed\r\n\r\n* Testing the auto-build functionality\r\n* Update examples by @snakers4 in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F137\r\n* Fx ssml and model loading by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F140\r\n* Update README.md by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F138\r\n* Tts v3 by @Islanna in https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fpull\u002F141\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fcompare\u002Fv0.1...v0.2","2022-05-23T16:37:05",{"id":197,"version":198,"summary_zh":199,"released_at":200},101543,"v0.1","This is a test release to test Github Action based PyPI publishing.\r\nSome proper semantic version will be created later.","2022-02-28T11:08:17",{"id":202,"version":203,"summary_zh":204,"released_at":205},101544,"v1","![header)](https:\u002F\u002Fuser-images.githubusercontent.com\u002F12515440\u002F89997349-b3523080-dc94-11ea-9906-ca2e8bc50535.png)\r\n\r\n [![Mailing list : test](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEmail-gray.svg?style=for-the-badge&logo=gmail)](mailto:hello@silero.ai) [![Mailing list : test](http:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTelegram-blue.svg?style=for-the-badge&logo=telegram)](https:\u002F\u002Ft.me\u002Fjoinchat\u002FBv9tjhpdXTI22OUgpOIIDg)\r\n[![License: CC BY-NC 4.0](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-GNU%20AGPL%203.0-lightgrey.svg?style=for-the-badge)](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002FLICENSE)\r\n\r\n\r\nWe publish the following models in this release:\r\n\r\n- English V1\r\n- German V1\r\n- Spanish V1\r\n\r\n| | PyTorch | ONNX | TensorFlow | Quantization | Quality | Colab | \r\n|-----------------|--------------------|--------------------|--------------------|--------------|---------|-------| \r\n| English (en_v1) | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | :hourglass: | [link](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FQuality-Benchmarks#latest) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples.ipynb) |\r\n| German (de_v1) | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | :hourglass: | [link](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FQuality-Benchmarks#latest) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples.ipynb) |\r\n| Spanish (es_v1) | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | :hourglass: | [link](https:\u002F\u002Fgithub.com\u002Fsnakers4\u002Fsilero-models\u002Fwiki\u002FQuality-Benchmarks#latest) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsnakers4\u002Fsilero-models\u002Fblob\u002Fmaster\u002Fexamples.ipynb) |","2020-09-16T12:20:23"]