[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-TEN-framework--ten-turn-detection":3,"tool-TEN-framework--ten-turn-detection":64},[4,17,25,39,48,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},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",[13,14,15],"开发框架","Agent","语言模型","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":10,"last_commit_at":23,"category_tags":24,"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",[13,15],{"id":26,"name":27,"github_repo":28,"description_zh":29,"stars":30,"difficulty_score":10,"last_commit_at":31,"category_tags":32,"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",[33,34,35,36,14,37,15,13,38],"图像","数据工具","视频","插件","其他","音频",{"id":40,"name":41,"github_repo":42,"description_zh":43,"stars":44,"difficulty_score":45,"last_commit_at":46,"category_tags":47,"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。",77062,3,"2026-04-04T04:44:48",[14,33,13,15,37],{"id":49,"name":50,"github_repo":51,"description_zh":52,"stars":53,"difficulty_score":45,"last_commit_at":54,"category_tags":55,"status":16},519,"PaddleOCR","PaddlePaddle\u002FPaddleOCR","PaddleOCR 是一款基于百度飞桨框架开发的高性能开源光学字符识别工具包。它的核心能力是将图片、PDF 等文档中的文字提取出来，转换成计算机可读取的结构化数据，让机器真正“看懂”图文内容。\n\n面对海量纸质或电子文档，PaddleOCR 解决了人工录入效率低、数字化成本高的问题。尤其在人工智能领域，它扮演着连接图像与大型语言模型（LLM）的桥梁角色，能将视觉信息直接转化为文本输入，助力智能问答、文档分析等应用场景落地。\n\nPaddleOCR 适合开发者、算法研究人员以及有文档自动化需求的普通用户。其技术优势十分明显：不仅支持全球 100 多种语言的识别，还能在 Windows、Linux、macOS 等多个系统上运行，并灵活适配 CPU、GPU、NPU 等各类硬件。作为一个轻量级且社区活跃的开源项目，PaddleOCR 既能满足快速集成的需求，也能支撑前沿的视觉语言研究，是处理文字识别任务的理想选择。",74913,"2026-04-05T10:44:17",[15,33,13,37],{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":45,"last_commit_at":62,"category_tags":63,"status":16},2181,"OpenHands","OpenHands\u002FOpenHands","OpenHands 是一个专注于 AI 驱动开发的开源平台，旨在让智能体（Agent）像人类开发者一样理解、编写和调试代码。它解决了传统编程中重复性劳动多、环境配置复杂以及人机协作效率低等痛点，通过自动化流程显著提升开发速度。\n\n无论是希望提升编码效率的软件工程师、探索智能体技术的研究人员，还是需要快速原型验证的技术团队，都能从中受益。OpenHands 提供了灵活多样的使用方式：既可以通过命令行（CLI）或本地图形界面在个人电脑上轻松上手，体验类似 Devin 的流畅交互；也能利用其强大的 Python SDK 自定义智能体逻辑，甚至在云端大规模部署上千个智能体并行工作。\n\n其核心技术亮点在于模块化的软件智能体 SDK，这不仅构成了平台的引擎，还支持高度可组合的开发模式。此外，OpenHands 在 SWE-bench 基准测试中取得了 77.6% 的优异成绩，证明了其解决真实世界软件工程问题的能力。平台还具备完善的企业级功能，支持与 Slack、Jira 等工具集成，并提供细粒度的权限管理，适合从个人开发者到大型企业的各类用户场景。",70612,"2026-04-05T11:12:22",[15,14,13,36],{"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":79,"owner_twitter":79,"owner_website":79,"owner_url":80,"languages":81,"stars":86,"forks":87,"last_commit_at":88,"license":89,"difficulty_score":45,"env_os":90,"env_gpu":90,"env_ram":90,"env_deps":91,"category_tags":96,"github_topics":97,"view_count":45,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":104,"updated_at":105,"faqs":106,"releases":142},442,"TEN-framework\u002Ften-turn-detection","ten-turn-detection","Turn detection for full-duplex dialogue communication","ten-turn-detection 是 TEN 开源框架下的核心组件，专门用于全双工对话通信中的“轮次检测”。通俗来讲，它赋予了 AI “听懂对话节奏”的能力，使其能够精准判断用户何时结束发言、何时停顿或是想要插话，从而实现像人类打电话一样自然流畅的双向语音交互。\n\n在开发语音 AI 代理或智能助手时，最大的挑战之一就是如何避免“抢话”或“反应迟钝”。ten-turn-detection 正是为了解决这一痛点，它能够有效区分长停顿与句子结束，支持全双工模式下的即时响应，极大地提升了语音交互的拟人化程度。\n\n该项目非常适合正在构建语音对话系统、实时通信应用或智能语音硬件的开发者与研究人员。它不仅提供了预置的高质量数据集，还公开了模型权重，方便用户快速部署与测试。作为 TEN 生态系统的一部分，ten-turn-detection 凭借其优异的检测性能，为构建低延迟、高响应速度的下一代语音 AI 应用提供了坚实的技术底座。","\u003Cdiv align=\"center\">\n\n![Image](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_f39b37a8c5a8.png)\n\n[![Discussion posts](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdiscussions\u002FTEN-framework\u002Ften-turn-detection?labelColor=gray&color=%20%23f79009)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fdiscussions\u002F)\n[![Commits](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcommit-activity\u002Fm\u002FTEN-framework\u002Ften-turn-detection?labelColor=gray&color=pink)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fgraphs\u002Fcommit-activity)\n[![Issues closed](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-search?query=repo%3ATEN-framework%2Ften-turn-detection%20is%3Aclosed&label=issues%20closed&labelColor=gray&color=green)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues)\n![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002Ften-framework\u002Ften-turn-detection?color=c4f042&labelColor=gray&style=flat-square)\n[![PRs Welcome](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPRs-welcome!-brightgreen.svg?style=flat-square)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fpulls)\n[![Hugging Face Space](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHugging%20Face-TEN%20Turn%20Detection-yellow?style=flat&logo=huggingface)](https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection)\n[![GitHub license](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache_2.0_with_certain_conditions-blue.svg?labelColor=%20%23155EEF&color=%20%23528bff)](.\u002FLICENSE)\n[![Ask DeepWiki](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002FTEN-framework\u002FTEN-turn-detection)\n\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n## Table of Contents\n\n- [Welcome to TEN](#welcome-to-ten)\n- [Introduction](#introduction)\n- [TEN Hugging Face Space](#ten-hugging-face-space)\n- [Key Features](#key-features)\n- [Prepared Dataset](#prepared-dataset)\n- [Detection Performance](#detection-performance)\n- [Quick Start](#quick-start)\n  - [Installation](#installation)\n  - [Model Weights](#model-weights)\n  - [Inference](#inference)\n- [Citation](#citation)\n- [TEN Ecosystem](#ten-ecosystem)\n- [Ask Questions](#ask-questions)\n- [License](#license)\n\n\u003Cbr>\n\n## Welcome to TEN\n\nTEN is an open-source framework for conversational voice AI agents.\n\n[TEN Ecosystem](#ten-ecosystem) includes [TEN Framework](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-framework), [Agent Examples](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples), [VAD](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-vad), [Turn Detection](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-turn-detection) and [Portal](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Fportal).\n\u003Cbr>\n\n| Community Channel | Purpose |\n| ---------------- | ------- |\n| [![Follow on X](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002FTenFramework?logo=X&color=%20%23f5f5f5)](https:\u002F\u002Ftwitter.com\u002Fintent\u002Ffollow?screen_name=TenFramework) | Follow TEN Framework on X for updates and announcements |\n| [![Follow on LinkedIn](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_8099e5fd328c.png)](https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Ften-framework) | Follow TEN Framework on LinkedIn for updates and announcements |\n| [![Discord TEN Community](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join%20TEN%20Community-5865F2?style=flat&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FVnPftUzAMJ) | Join our Discord community to connect with developers |\n| [![Hugging Face Space](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHugging%20Face-TEN%20Framework-yellow?style=flat&logo=huggingface)](https:\u002F\u002Fhuggingface.co\u002FTEN-framework) | Join our Hugging Face community to explore our spaces and models |\n| [![WeChat](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTEN_Framework-WeChat_Group-%2307C160?logo=wechat&labelColor=darkgreen&color=gray)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-agent\u002Fdiscussions\u002F170) | Join our WeChat group for Chinese community discussions |\n\n\u003Cbr>\n\n> \\[!IMPORTANT]\n>\n> **Star TEN Repositories** ⭐️\n>\n> Get instant notifications for new releases and updates. Your support helps us grow and improve TEN!\n\n\u003Cbr>\n\n![TEN star us gif](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_5a5e215edc7c.png)\n\n\u003Cbr>\n\n## TEN Hugging Face Space\n\n\u003Chttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F725a8318-d679-4b17-b9e4-e3dce999b298>\n\nYou are more than welcome to [Visit TEN Hugging Face Space](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTEN-framework\u002Ften-agent-demo) to try VAD and Turn Detection together.\n\n## Introduction\n\n**TEN Turn Detection** is an advanced intelligent turn detection model designed specifically for natural and dynamic communication between humans and AI agents. This technology addresses one of the most challenging aspects of human-AI conversation: detecting natural turn-taking cues and enabling contextually-aware interruptions. TEN Turn Detection incorporates deep semantic understanding of conversation context and linguistic patterns to create more natural dialogue with AI.\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"images\u002Fturn_detection.svg\" alt=\"TEN Turn Detection SVG Diagram\" width=\"800\"\u002F>\n\u003C\u002Fdiv>\n\n**TEN Turn Detection** categorizes user's text into three key states:\n\n**finished**: A finished utterance where the user has expressed a complete thought and expects a response. Example: \"Hey there I was wondering can you help me with my order\"\n\n**wait**: An wait utterance where the user has explicitly instructed the AI not to speak. Example: \"Shut up\"\n\n**unfinished**: A clearly unfinished utterance where the user has momentarily paused but intends to continue speaking. Example: \"Hello I have a question about\"\n\nThese three classification states allow the TEN system to create natural conversation dynamics by intelligently managing turn-taking, reducing awkward interruptions while maintaining conversation flow.\n\nTEN Turn Detection utilizes a multi-layered approach based on the transformer-based language model(Qwen2.5-7B) for semantic analysis.\n\n## Key Features\n\n- **Context-Aware Turn Management**\n  TEN Turn Detection analyzes linguistic patterns and semantic context to accurately identify turn completion points. This capability enables intelligent interruption handling, allowing the system to determine when interruptions are contextually appropriate while maintaining natural conversation flow across various dialogue scenarios.\n\n- **Multilingual Turn Detection Support**\n  TEN Turn Detection provides comprehensive support for both English and Chinese languages. It is engineered to accurately identify turn-taking cues and completion signals across multilingual conversations.\n\n- **Superior Performance**\n  Compared with multiple open-source solutions, TEN achieves superior performance across all metrics on our publicly available test dataset.\n\n## Prepared Dataset\n\nWe have open-sourced the TEN-Turn-Detection TestSet, a bilingual (Chinese and English) collection of conversational inputs specifically designed to evaluate turn detection capabilities in AI dialogue systems. The dataset consists of three distinct components:\n\n_wait.txt_: Contains expressions requesting conversation pauses or termination\n\n_unfinished.txt_: Features incomplete dialogue inputs with truncated utterances\n\n_finished.txt_: Provides complete conversational inputs across multiple domains\n\n## Detection Performance\n\nWe conducted comprehensive evaluations comparing several open-source models for turn detection using our test dataset:\n\n\u003Cdiv align=\"center\">\n\n| LANGUAGE |         MODEL          | FINISHED\u003Cbr>ACCURACY | UNFINISHED\u003Cbr>ACCURACY | WAIT\u003Cbr>ACCURACY |\n| :------: | :--------------------: | :------------------: | :--------------------: | :--------------: |\n| English  |        Model A         |        59.74%        |         86.46%         |       N\u002FA        |\n| English  |        Model B         |        71.61%        |         96.88%         |       N\u002FA        |\n| English  | **TEN Turn Detection** |      **90.64%**      |       **98.44%**       |     **91%**      |\n\n| LANGUAGE |         MODEL          | FINISHED\u003Cbr>ACCURACY | UNFINISHED\u003Cbr>ACCURACY | WAIT\u003Cbr>ACCURACY |\n| :------: | :--------------------: | :------------------: | :--------------------: | :--------------: |\n| Chinese  |        Model B         |        74.63%        |         88.89%         |       N\u002FA        |\n| Chinese  | **TEN Turn Detection** |      **98.90%**      |       **92.74%**       |     **92%**      |\n\n\u003C\u002Fdiv>\n\n> **Notes:**\n>\n> 1. Model A doesn't support Chinese language processing\n> 2. Neither Model A nor Model B support the \"WAIT\" state detection\n\n## Quick Start\n\n### Installation\n\n```\ngit clone https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection.git\npip install \"transformers>=4.45.0\"\npip install \"torch>=2.0.0\"\n```\n\n### Model Weights\n\nThe TEN Turn Detection model is available on HuggingFace:\n\n- Model Repository: [TEN-framework\u002FTEN_Turn_Detection](https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection)\n\nYou can download the model in several ways:\n\n1. **Automatic download** (recommended): The model weights will be automatically downloaded when you run the inference script for the first time. HuggingFace Transformers will cache the model locally.\n\n2. **Using Git LFS**:\n\n   ```bash\n   # Install Git LFS if you haven't already\n   git lfs install\n\n   # Clone the repository with model weights\n   git clone https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection\n   ```\n\n3. **Using the Hugging Face Hub library**:\n\n   ```python\n   from huggingface_hub import snapshot_download\n\n   snapshot_download(repo_id=\"TEN-framework\u002FTEN_Turn_Detection\")\n   ```\n\n### Inference\n\nThe inference script accepts command line arguments for user input:\n\n```\n# Basic usage\npython inference.py --input \"Your text to analyze\"\n\n```\n\nExample output:\n\n```\nLoading model from TEN-framework\u002FTEN_Turn_Detection...\nRunning inference on: 'Hello I have a question about'\n\nResults:\nInput: 'Hello I have a question about'\nTurn Detection Result: 'unfinished'\n```\n\n## Citation\n\nIf you use TEN Turn Detection in your research or applications, please cite:\n\n```\n@misc{TEN_Turn_Detection,\nauthor = {TEN Team},\ntitle = {TEN Turn Detection: Turn detection for full-duplex dialogue communication\n\n},\nyear = {2025},\nurl = {https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection},\n}\n```\n\n## TEN Ecosystem\n\n| Project | Preview |\n| ------- | ------- |\n| [**️TEN Framework**][ten-framework-link]\u003Cbr>Open-source framework for conversational AI Agents.\u003Cbr>\u003Cbr>![][ten-framework-shield] | ![][ten-framework-banner] |\n| [**TEN VAD**][ten-vad-link]\u003Cbr>Low-latency, lightweight and high-performance streaming voice activity detector (VAD).\u003Cbr>\u003Cbr>![][ten-vad-shield] | ![][ten-vad-banner] |\n| [**️ TEN Turn Detection**][ten-turn-detection-link]\u003Cbr>TEN Turn Detection enables full-duplex dialogue communication.\u003Cbr>\u003Cbr>![][ten-turn-detection-shield] | ![][ten-turn-detection-banner] |\n| [**TEN Agent Examples**][ten-agent-example-link]\u003Cbr>Usecases powered by TEN.\u003Cbr>\u003Cbr> | ![][ten-agent-example-banner] |\n| [**TEN Portal**][ten-portal-link]\u003Cbr>The official site of the TEN Framework with documentation and a blog.\u003Cbr>\u003Cbr>![][ten-portal-shield] | ![][ten-portal-banner] |\n\n\n\u003Cbr>\n\n## Ask Questions\n\n[![Ask DeepWiki](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002FTEN-framework\u002FTEN-framework)\n\nMost questions can be answered by using DeepWiki, it is fast, intutive to use and supports multiple languages.\n\n\u003Cbr>\n\n## License\n\nThis project is released pursuant to the Apache License, Version 2.0, with additional restrictions.For details, please refer to the[ LICENSE](.\u002FLICENSE) file.\n\n[ten-framework-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften_framework?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-framework-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2a560a74-68f3-4f4a-9ec8-89464c42a9c7\n[ten-framework-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften_framework\n\n[ten-vad-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-vad\n[ten-vad-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften-vad?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-vad-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe504135e-67fd-4fa1-b0e4-d495358d8aa5\n\n[ten-turn-detection-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-turn-detection\n[ten-turn-detection-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften-turn-detection?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-turn-detection-banner]: https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_f39b37a8c5a8.png\n\n[ten-agent-example-link]: https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples\n[ten-agent-example-banner]:https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F7f735633-c7f6-4432-b6b4-d2a2977ca588\n\n[ten-portal-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Fportal\n[ten-portal-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Fportal?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-portal-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Ff56c75b9-722c-4156-902d-ae98ce2b3b5e\n","\u003Cdiv align=\"center\">\n\n![Image](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_f39b37a8c5a8.png)\n\n[![Discussion posts](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdiscussions\u002FTEN-framework\u002Ften-turn-detection?labelColor=gray&color=%20%23f79009)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fdiscussions\u002F)\n[![Commits](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcommit-activity\u002Fm\u002FTEN-framework\u002Ften-turn-detection?labelColor=gray&color=pink)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fgraphs\u002Fcommit-activity)\n[![Issues closed](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-search?query=repo%3ATEN-framework%2Ften-turn-detection%20is%3Aclosed&label=issues%20closed&labelColor=gray&color=green)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues)\n![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002Ften-framework\u002Ften-turn-detection?color=c4f042&labelColor=gray&style=flat-square)\n[![PRs Welcome](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPRs-welcome!-brightgreen.svg?style=flat-square)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fpulls)\n[![Hugging Face Space](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHugging%20Face-TEN%20Turn%20Detection-yellow?style=flat&logo=huggingface)](https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection)\n[![GitHub license](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache_2.0_with_certain_conditions-blue.svg?labelColor=%20%23155EEF&color=%20%23528bff)](.\u002FLICENSE)\n[![Ask DeepWiki](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002FTEN-framework\u002FTEN-turn-detection)\n\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n## 目录\n\n- [欢迎来到 TEN](#欢迎来到-ten)\n- [简介](#简介)\n- [TEN Hugging Face Space](#ten-hugging-face-space)\n- [主要特性](#主要特性)\n- [准备好的数据集](#准备好的数据集)\n- [检测性能](#检测性能)\n- [快速开始](#快速开始)\n  - [安装](#安装)\n  - [模型权重](#模型权重)\n  - [推理](#推理)\n- [引用](#引用)\n- [TEN 生态系统](#ten-生态系统)\n- [提问](#提问)\n- [许可证](#许可证)\n\n\u003Cbr>\n\n## 欢迎来到 TEN\n\nTEN 是一个用于对话式语音 AI Agent（智能体）的开源框架。\n\n[TEN 生态系统](#ten-生态系统) 包括 [TEN Framework](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-framework)、[Agent Examples（智能体示例）](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples)、[VAD（Voice Activity Detection，语音活动检测）](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-vad)、[Turn Detection（轮次检测）](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-turn-detection) 和 [Portal](https:\u002F\u002Fgithub.com\u002Ften-framework\u002Fportal)。\n\u003Cbr>\n\n| 社区渠道 | 用途 |\n| ---------------- | ------- |\n| [![Follow on X](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002FTenFramework?logo=X&color=%20%23f5f5f5)](https:\u002F\u002Ftwitter.com\u002Fintent\u002Ffollow?screen_name=TenFramework) | 在 X 上关注 TEN Framework 以获取更新和公告 |\n| [![Follow on LinkedIn](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_8099e5fd328c.png)](https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Ften-framework) | 在 LinkedIn 上关注 TEN Framework 以获取更新和公告 |\n| [![Discord TEN Community](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join%20TEN%20Community-5865F2?style=flat&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FVnPftUzAMJ) | 加入我们的 Discord 社区与开发者交流 |\n| [![Hugging Face Space](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHugging%20Face-TEN%20Framework-yellow?style=flat&logo=huggingface)](https:\u002F\u002Fhuggingface.co\u002FTEN-framework) | 加入我们的 Hugging Face 社区，探索我们的 Spaces 和模型 |\n| [![WeChat](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTEN_Framework-WeChat_Group-%2307C160?logo=wechat&labelColor=darkgreen&color=gray)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-agent\u002Fdiscussions\u002F170) | 加入我们的微信群进行中文社区讨论 |\n\n\u003Cbr>\n\n> \\[!IMPORTANT]\n>\n> **点亮 TEN 仓库 Star** ⭐️\n>\n> 获取新版本和更新的即时通知。您的支持帮助我们成长和改进 TEN！\n\n\u003Cbr>\n\n![TEN star us gif](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_5a5e215edc7c.png)\n\n\u003Cbr>\n\n## TEN Hugging Face Space\n\n\u003Chttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F725a8318-d679-4b17-b9e4-e3dce999b298>\n\n非常欢迎您 [访问 TEN Hugging Face Space](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTEN-framework\u002Ften-agent-demo) 体验 VAD 和 Turn Detection 的结合。\n\n## 简介\n\n**TEN Turn Detection** 是一个先进的智能轮次检测模型，专为人类与 AI Agent 之间自然、动态的交流而设计。该技术解决了人机对话中最具挑战性的问题之一：检测自然的轮次转换线索并启用具备上下文感知的中断机制。TEN Turn Detection 融合了对对话上下文和语言模式的深度语义理解，旨在与 AI 创造更自然的对话体验。\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"images\u002Fturn_detection.svg\" alt=\"TEN Turn Detection SVG Diagram\" width=\"800\"\u002F>\n\u003C\u002Fdiv>\n\n**TEN Turn Detection** 将用户的文本归类为三种关键状态：\n\n**finished（已完成）**：用户已表达完整想法并期待回应的完整话语。示例：\"Hey there I was wondering can you help me with my order\"\n\n**wait（等待）**：用户明确指示 AI 不要说话的等待话语。示例：\"Shut up\"\n\n**unfinished（未完成）**：明显未完成的话语，用户暂时停顿但打算继续说话。示例：\"Hello I have a question about\"\n\n这三种分类状态使 TEN 系统能够通过智能管理轮次转换来创造自然的对话动态，在保持对话流畅的同时减少尴尬的中断。\n\nTEN Turn Detection 利用基于 Transformer（变换器）的语言模型（Qwen2.5-7B）的多层方法进行语义分析。\n\n## 主要特性\n\n- **上下文感知的轮次管理**\n  TEN Turn Detection 分析语言模式和语义上下文，以准确识别轮次完成点。此功能实现了智能中断处理，允许系统在各种对话场景中确定何时中断是符合上下文情境的，同时保持自然的对话流畅度。\n\n- **多语言轮次检测支持**\n  TEN Turn Detection 为英语和中文提供全面支持。它旨在准确识别多语言对话中的轮次转换线索和完成信号。\n\n- **卓越的性能**\n  与多种开源解决方案相比，TEN 在我们公开的测试数据集上的所有指标均实现了卓越的性能。\n\n## 已准备的数据集\n\n我们开源了 TEN-Turn-Detection 测试集，这是一个专为评估 AI 对话系统中的轮次检测（Turn Detection）能力而设计的双语（中文和英文）对话输入集合。该数据集包含三个不同的组成部分：\n\n_wait.txt_：包含请求暂停或终止对话的表达\n\n_unfinished.txt_：包含不完整的对话输入，话语被截断\n\n_finished.txt_：提供跨多个领域的完整对话输入\n\n## 检测性能\n\n我们使用我们的测试数据集进行了全面的评估，比较了几个用于轮次检测的开源模型：\n\n\u003Cdiv align=\"center\">\n\n| 语言 |         模型          | FINISHED\u003Cbr>准确率 | UNFINISHED\u003Cbr>准确率 | WAIT\u003Cbr>准确率 |\n| :------: | :--------------------: | :------------------: | :--------------------: | :--------------: |\n| English  |        Model A         |        59.74%        |         86.46%         |       N\u002FA        |\n| English  |        Model B         |        71.61%        |         96.88%         |       N\u002FA        |\n| English  | **TEN Turn Detection** |      **90.64%**      |       **98.44%**       |     **91%**      |\n\n| 语言 |         模型          | FINISHED\u003Cbr>准确率 | UNFINISHED\u003Cbr>准确率 | WAIT\u003Cbr>准确率 |\n| :------: | :--------------------: | :------------------: | :--------------------: | :--------------: |\n| Chinese  |        Model B         |        74.63%        |         88.89%         |       N\u002FA        |\n| Chinese  | **TEN Turn Detection** |      **98.90%**      |       **92.74%**       |     **92%**      |\n\n\u003C\u002Fdiv>\n\n> **注意：**\n>\n> 1. Model A 不支持中文处理\n> 2. Model A 和 Model B 均不支持 \"WAIT\" 状态检测\n\n## 快速开始\n\n### 安装\n\n```\ngit clone https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection.git\npip install \"transformers>=4.45.0\"\npip install \"torch>=2.0.0\"\n```\n\n### 模型权重\n\nTEN Turn Detection 模型已在 HuggingFace 上发布：\n\n- 模型仓库：[TEN-framework\u002FTEN_Turn_Detection](https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection)\n\n您可以通过以下几种方式下载模型：\n\n1. **自动下载**（推荐）：当您首次运行推理脚本时，模型权重将自动下载。HuggingFace Transformers 会将模型缓存在本地。\n\n2. **使用 Git LFS**：\n\n   ```bash\n   # Install Git LFS if you haven't already\n   git lfs install\n\n   # Clone the repository with model weights\n   git clone https:\u002F\u002Fhuggingface.co\u002FTEN-framework\u002FTEN_Turn_Detection\n   ```\n\n3. **使用 Hugging Face Hub 库**：\n\n   ```python\n   from huggingface_hub import snapshot_download\n\n   snapshot_download(repo_id=\"TEN-framework\u002FTEN_Turn_Detection\")\n   ```\n\n### 推理\n\n推理脚本接受用户输入的命令行参数：\n\n```\n# Basic usage\npython inference.py --input \"Your text to analyze\"\n\n```\n\n示例输出：\n\n```\nLoading model from TEN-framework\u002FTEN_Turn_Detection...\nRunning inference on: 'Hello I have a question about'\n\nResults:\nInput: 'Hello I have a question about'\nTurn Detection Result: 'unfinished'\n```\n\n## 引用\n\n如果您在研究或应用中使用 TEN Turn Detection，请引用：\n\n```\n@misc{TEN_Turn_Detection,\nauthor = {TEN Team},\ntitle = {TEN Turn Detection: Turn detection for full-duplex dialogue communication\n\n},\nyear = {2025},\nurl = {https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection},\n}\n```\n\n## TEN 生态系统\n\n| 项目 | 预览 |\n| ------- | ------- |\n| [**️TEN Framework**][ten-framework-link]\u003Cbr>用于对话式 AI Agent 的开源框架。\u003Cbr>\u003Cbr>![][ten-framework-shield] | ![][ten-framework-banner] |\n| [**TEN VAD**][ten-vad-link]\u003Cbr>低延迟、轻量级且高性能的流式语音活动检测器 (VAD)。\u003Cbr>\u003Cbr>![][ten-vad-shield] | ![][ten-vad-banner] |\n| [**️ TEN Turn Detection**][ten-turn-detection-link]\u003Cbr>TEN Turn Detection 实现全双工对话通信。\u003Cbr>\u003Cbr>![][ten-turn-detection-shield] | ![][ten-turn-detection-banner] |\n| [**TEN Agent Examples**][ten-agent-example-link]\u003Cbr>基于 TEN 的用例。\u003Cbr>\u003Cbr> | ![][ten-agent-example-banner] |\n| [**TEN Portal**][ten-portal-link]\u003Cbr>TEN Framework 的官方网站，包含文档和博客。\u003Cbr>\u003Cbr>![][ten-portal-shield] | ![][ten-portal-banner] |\n\n\n\u003Cbr>\n\n## 提问\n\n[![Ask DeepWiki](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002FTEN-framework\u002FTEN-framework)\n\n大多数问题可以通过使用 DeepWiki 得到解答，它速度快、直观易用且支持多种语言。\n\n\u003Cbr>\n\n## 许可证\n\n本项目根据 Apache License 2.0 版本发布，并包含额外限制。详情请参阅 [LICENSE](.\u002FLICENSE) 文件。\n\n[ten-framework-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften_framework?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-framework-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2a560a74-68f3-4f4a-9ec8-89464c42a9c7\n[ten-framework-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften_framework\n\n[ten-vad-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-vad\n[ten-vad-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften-vad?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-vad-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe504135e-67fd-4fa1-b0e4-d495358d8aa5\n\n[ten-turn-detection-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Ften-turn-detection\n[ten-turn-detection-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Ften-turn-detection?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-turn-detection-banner]: https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_readme_f39b37a8c5a8.png\n\n[ten-agent-example-link]: https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples\n[ten-agent-example-banner]:https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F7f735633-c7f6-4432-b6b4-d2a2977ca588\n\n[ten-portal-link]: https:\u002F\u002Fgithub.com\u002Ften-framework\u002Fportal\n[ten-portal-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ften-framework\u002Fportal?color=ffcb47&labelColor=gray&style=flat-square&logo=github\n[ten-portal-banner]: https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Ff56c75b9-722c-4156-902d-ae98ce2b3b5e","# TEN Turn Detection 快速上手指南\n\nTEN Turn Detection 是一个基于 Qwen2.5-7B 的高级对话轮次检测模型，支持中英文，能够精准识别“已完成”、“未完成”和“等待”三种对话状态，适用于构建自然的 AI 语音交互系统。\n\n## 环境准备\n\n在开始之前，请确保您的开发环境满足以下要求：\n\n*   **Python**: 建议使用 Python 3.8 或更高版本。\n*   **深度学习框架**:\n    *   `torch >= 2.0.0`\n    *   `transformers >= 4.45.0`\n*   **硬件建议**: 模型基于 Qwen2.5-7B，建议配备具有足够显存（如 16GB+）的 GPU 以获得最佳推理性能。\n\n## 安装步骤\n\n1.  **克隆项目代码**\n    从 GitHub 获取项目源码：\n\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection.git\n    ```\n\n2.  **安装依赖库**\n    安装必要的 Python 依赖包：\n\n    ```bash\n    pip install \"transformers>=4.45.0\"\n    pip install \"torch>=2.0.0\"\n    ```\n\n## 基本使用\n\n### 1. 准备模型权重\n模型权重托管在 HuggingFace 上（`TEN-framework\u002FTEN_Turn_Detection`）。\n推荐方式为**自动下载**：首次运行推理脚本时，`transformers` 库会自动从 HuggingFace 下载并缓存模型权重到本地。\n\n### 2. 运行推理\n使用项目提供的 `inference.py` 脚本进行文本状态检测。\n\n**基本命令：**\n\n```bash\npython inference.py --input \"Your text to analyze\"\n```\n\n**示例：检测未完成语句**\n\n输入一段明显未说完的话：\n\n```bash\npython inference.py --input \"Hello I have a question about\"\n```\n\n**预期输出结果：**\n\n```text\nLoading model from TEN-framework\u002FTEN_Turn_Detection...\nRunning inference on: 'Hello I have a question about'\n\nResults:\nInput: 'Hello I have a question about'\nTurn Detection Result: 'unfinished'\n```\n\n模型将返回以下三种状态之一：\n*   `finished`: 用户表达完整，期待回复。\n*   `unfinished`: 用户暂停但意图继续说话。\n*   `wait`: 用户明确指示 AI 保持静默（如“闭嘴”）。","某智能客服团队正在开发一款语音交互机器人，用户需要通过语音描述复杂的账户问题，机器人需要准确判断何时应该接话或回答。\n\n### 没有 ten-turn-detection 时\n\n- 用户说话时稍微停顿思考，机器人就误以为用户说完了，突然插话打断用户思路，交互体验非常生硬。\n- 为了避免误打断，开发者只能被迫设置较长的静音等待时间（如 1.5 秒），导致用户说完话后机器人反应迟钝，对话有明显的“卡顿感”。\n- 当用户想要打断机器人进行纠正或补充时，机器人无法识别用户的抢话意图，只能自顾自地把预设话术讲完，显得不够智能。\n- 传统的 VAD（语音活动检测）仅能判断“有没有声音”，无法区分“嗯...”等语气词和真正的语句结束，导致对话状态管理混乱。\n\n### 使用 ten-turn-detection 后\n\n- ten-turn-detection 能够精准区分“思考停顿”与“表达结束”，用户在思考时机器人会耐心等待，不再出现尴尬的抢话现象。\n- 模型能够预测对话轮次的结束点，大幅缩短了不必要的静音等待时间，实现了毫秒级的响应速度，对话像真人一样流畅自然。\n- 支持全双工通信模式，当用户试图打断机器人时，系统能立即检测到新轮次的开始并停止播放，实现了自然的“抢话”交互。\n- 基于语义理解进行检测，有效过滤语气词和呼吸声等噪音，让机器人对“何时开口”的判断更加智能准确。\n\nten-turn-detection 通过精准的轮次检测能力，完美解决了语音交互中“抢话”与“反应慢”的核心矛盾，让 AI 对话真正实现了类人的自然流畅。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FTEN-framework_ten-turn-detection_f39b37a8.jpg","TEN-framework","TEN framework","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FTEN-framework_23fbc0fc.png","",null,"https:\u002F\u002Fgithub.com\u002FTEN-framework",[82],{"name":83,"color":84,"percentage":85},"Python","#3572A5",100,545,35,"2026-04-03T22:30:47","NOASSERTION","未说明",{"notes":92,"python":90,"dependencies":93},"README 中未明确列出操作系统、硬件及 Python 版本要求。根据模型介绍，底层基于 Qwen2.5-7B，建议配备高性能 GPU（如显存 16GB 以上）以保证推理性能。模型权重托管于 HuggingFace，首次运行会自动下载。",[94,95],"transformers>=4.45.0","torch>=2.0.0",[15],[98,99,100,101,102,103],"large-language-models","natural-language-understanding","real-time","turn-taking","turn-detection","turn-detector","2026-03-27T02:49:30.150509","2026-04-06T06:45:55.812449",[107,112,117,122,127,132,137],{"id":108,"question_zh":109,"answer_zh":110,"source_url":111},1701,"是否有更小、更轻量的模型版本适合实时检测？","有的。除了 7B 版本外，社区维护者提供了一个基于 Qwen 3 0.6B 训练的轻量级版本，更适合对延迟要求较高的实时场景。模型地址：https:\u002F\u002Fhuggingface.co\u002Fdangvansam\u002FQwen3-0.6B-turn-detection-en","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F23",{"id":113,"question_zh":114,"answer_zh":115,"source_url":116},1702,"如何在多轮对话中使用上下文进行轮次检测？","项目说明中的“上下文感知”指的是单轮输入文本内部的语言上下文，而非多轮对话的历史记录。目前的示例代码主要针对单轮处理，维护者表示单轮处理方式已能有效应对大多数场景，通常无需传入多轮对话历史。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F17",{"id":118,"question_zh":119,"answer_zh":120,"source_url":121},1703,"运行推理时遇到 TokenizerFast 错误如何解决？","该错误通常是由于 `transformers` 或 `tokenizers` 库版本不匹配导致的。请确保安装了以下版本或更高版本：`transformers >= 4.45.0` 和 `tokenizers >= 0.20.3`。升级命令：`pip install --upgrade \"transformers>=4.45.0\" \"tokenizers>=0.20.3\"`。如果问题依旧，可尝试删除 HuggingFace 缓存：`rm -rf ~\u002F.cache\u002Fhuggingface\u002Fhub`。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F9",{"id":123,"question_zh":124,"answer_zh":125,"source_url":126},1704,"项目是否支持 NodeJS 或 Web 开发集成？","目前官方尚未提供 NPM 包。建议的集成方式是在 Python 环境中运行模型推理，并通过 RESTful API 将检测结果暴露给 NodeJS 或其他 Web 服务调用。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F3",{"id":128,"question_zh":129,"answer_zh":130,"source_url":131},1705,"推理代码中的 Chat Template 处理方式有何特殊之处？","该模型将轮次检测视为文本分类任务而非文本补全任务。因此，在推理代码中保留了 `\u003C|im_end|>` 等标记作为输入的一部分，这与 Livekit 等将其视为文本补全任务的处理方式有所不同。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F21",{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},1706,"模型是否支持批量推理？","支持。你可以利用 `transformers` 库中的 `batch_decode` 方法来实现批量推理，以提高处理效率。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F28",{"id":138,"question_zh":139,"answer_zh":140,"source_url":141},1707,"在哪里可以下载项目的测试数据集？","测试数据集已在项目仓库中开源，可以直接在 GitHub 的 `TEN-Turn-TestSet` 目录下找到：https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Ftree\u002Fmain\u002FTEN-Turn-TestSet。","https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-turn-detection\u002Fissues\u002F29",[]]