NotebookLM新增深度研究功能,并支持更多源文件类型。

内容来源:https://blog.google/technology/google-labs/notebooklm-deep-research-file-types/
内容总结:
智能笔记工具NotebookLM近日推出深度研究功能并扩大文件兼容范围,以提升用户研究效率。该工具通过深度研究代理功能,可基于用户提出的问题自动生成研究计划,在数分钟内扫描数百个网络资源,最终输出结构清晰、来源可靠的综合分析报告。值得注意的是,生成报告及其引用来源可直接导入工作笔记,用户还能同步添加其他资料源,结合音频视频概览等功能构建完整知识体系。
在文件支持方面,本次更新新增对五大格式的兼容:用户现可直接导入谷歌表格进行数据统计与摘要生成;通过链接批量添加云端文件;上传手写笔记照片等图像资料;从谷歌云端硬盘直接调用PDF文档;以及分析微软Word文档内容。这些功能将在本周内向全体用户陆续开放,其中图像上传功能预计在未来几周内全面上线。此次升级旨在打破多格式资料处理壁垒,为学术研究和工作决策提供更高效的数字工具支持。
中文翻译:
NotebookLM推出深度研究功能并支持更多文件类型
可靠的信息源是所有研究项目的基石,也是NotebookLM的使用基础。今天我们推出多项新功能,助您更高效地发现和运用信息源:深度研究智能助手正式上线,同时新增多种日常文件格式支持。
用深度研究拓展信息维度
深度研究功能现已全面上线,可自动化处理复杂网络调研并简化研究流程。它如同您的专属研究员,既能生成深度研究报告,也能推荐相关文章、论文或网站——您还可以指定这位"研究员"在特定领域进行搜索。
在您的指引下,深度研究将根据问题制定研究计划,自动浏览数百个网站,并在学习过程中持续优化搜索策略。仅需几分钟,就能生成结构清晰、见解深刻且严格标注信息来源的研究报告。
NotebookLM深度研究的独特之处在于:报告仅是起点。您可将报告及其原始资料直接导入笔记,在后台持续运行研究任务的同时继续添加其他资料,无需中断工作流即可构建丰富的专题知识库。随后您还能运用音频/视频摘要等 NotebookLM 特色功能提炼核心观点,或通过内容转换深化对主题的理解。
深度研究使用指南:
• 启动搜索:在资料来源面板选择“网页”来源
• 选择研究模式:
• 快速研究:适合即时检索,可快速扫描信息并立即审阅导入资料
支持更多文件格式上传
您的工作涉及多种文件类型,研究工具也应如此。根据用户反馈,我们新增支持以下格式:
• Google Sheets:上传结构化数据,查询关键统计数据或生成电子表格摘要
• 网盘链接直传:通过复制粘贴即可添加云端文件链接,操作如同添加网页或YouTube视频。支持逗号分隔的多链接同时添加
• 图像文件:可上传手写笔记照片或宣传册图片
• 谷歌云端硬盘PDF:直接添加云端的研究论文、报告或电子书,省去下载重传步骤
• Microsoft Word文档(.docx):上传.docx文件用于分析笔记、草稿等文档
所有功能将在一周内向全体用户开放(图片上传功能将于数周内推出)。期待见证这些新功能如何助力您在NotebookLM中提升研究效率。
英文来源:
NotebookLM adds Deep Research and support for more source types
A strong set of trusted sources is the foundation of any research project, and the foundation of how you use NotebookLM. Today, NotebookLM is introducing new ways to help you find and use sources more effectively: using Deep Research agents and supporting more of the file types you use every day.
Expand your sources with Deep Research
We are now rolling out Deep Research to automate and simplify complex online research. It acts like your dedicated researcher, synthesising a detailed report or recommending relevant articles, papers or websites — and you can direct this "researcher" to search specific places, too.
With your guidance, Deep Research takes your question, creates a research plan and browses hundreds of websites on your behalf, refining its search as it learns. In a few minutes, it generates an organized, insightful, source-grounded report.
Here’s what makes NotebookLM's Deep Research different: the report is just the beginning. You can add the report and its sources directly into your notebook. You can continue to add other sources while Deep Research runs in the background, helping you assemble a rich knowledge base on any topic without leaving your workflow. And you can then use any of NotebookLM’s capabilities, such as Audio or Video overviews, to pull insights or transform content to understand the topic better.
Here’s how to try Deep Research in NotebookLM:
- Start your search: In the source panel, select Web as a source.
- Choose your research style:
- Fast Research: Best for a quick search. It rapidly scans for information, letting you immediately review and import sources.
Upload information in more file types
Your work spans multiple file types, and your research tool should, too. Based on your feedback, we’re adding support for new file types: - Google Sheets: Add structured data and ask information on key statistics or generate summaries from your spreadsheets.
- Drive files as URLs: Add drive files as URLs, with a simple copy-paste, just like any website or YouTube video. You can even add multiple links, separated with commas
- Images: Upload a photo of your hand-written notes or a brochure you got.
- PDFs from Google Drive: Add research papers, reports or ebooks directly from Drive, skipping downloading and re-uploading.
- Microsoft Word documents (.docx): Upload your .docx files to analyze notes, drafts and other documents.
All these features will be available to all users over the next week with images becoming available over the next few weeks. We’re looking forward to seeing how you’ll boost your research with all these new sources in NotebookLM.
文章标题:NotebookLM新增深度研究功能,并支持更多源文件类型。
文章链接:https://www.qimuai.cn/?post=2056
本站文章均为原创,未经授权请勿用于任何商业用途