lobehubbot
commented
5 months ago Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
🥰 Description of requirements
I hope to be able to support parsing files that can extract text, such as doc, pdf, txt, etc., and then put the extracted text directly into the context for simple dialogue with the file. Based on the long context common for large models today, this should be the best way to handle files that are not too large. And it should be possible to implement without a server-side database.
🧐 Solution
The chatnio project seems to be a similar solution, maybe you can refer to it.
📝 Supplementary information
None
xiaoxuan220
🥰 需求描述
我希望可以支持解析能提取文字的文件如doc、pdf、txt等，然后将提取到的文字直接放进上下文中进行简单的与文件对话。基于现在大模型普遍的长上下文，这应该是处理不太大的文件效果最好的方式。而且应该可以在没有服务端数据库的前提下实现。
🧐 解决方案
chatnio项目似乎是差不多的方案，或许可以参考一下。
📝 补充信息
无