ubos-tech / ubos-template-pinecone-bot-nr

Get Started • Community • Youtube • Discord • GitHub

Welcome to our "Open AI Telegram Bot + Pinecone" template! With the help of this template, you can effortlessly create a comprehensive knowledge base for your bot. It offers support for processing diverse types of data, including .txt and .pdf files, as well as the ability to read links and extract valuable text information from websites.

To enhance the functionality of the bot, we have integrated the Pinecone vector base, which serves as both a database and a semantic search engine. This enables efficient storage and retrieval of information based on similarity and relevance. Additionally, our template leverages Telegram and GPT chat for seamless user interaction.

To quickly get started

We recommend filling out the settings page. There, you will find all the necessary links for generating keys and configuring the template to suit your specific requirements. These settings will ensure smooth integration and optimal performance of your Open AI Telegram Bot + Pinecone.

How you can create your Knowledge base

1. First, you need to create your database

   

2. File Formats:

You can include files in .txt and .pdf formats to store textual information in your database. Ensure that the files you add comply with these formats for successful ingestion and processing. Internet Resource Links:

If you want to retrieve text information from a website, you can provide the link to the internet resource. The template will fetch the relevant text from the specified webpage and incorporate it into your knowledge base.

Language Requirements:

Currently, the "Open AI Telegram Bot + Pinecone" template supports English language content exclusively. Make sure that the files or webpages you add for text extraction are in English to ensure accurate processing and optimal results.

File name:

Each file must have a unique name.

3. To train your bot using the prepared knowledge base, follow these steps:

• Save Data in Pinecone: Save the data in Pinecone, ensuring it is organized and compatible with the platform.

• Choose Text Division: Select the preferred text division method (sentences or paragraphs) during the saving process.

• Edit Content of Vectors: After saving in Pinecone, you can modify the content of vectors but cannot change the text division type.

By training your bot with the knowledge base stored in Pinecone, it will be equipped to perform semantic searches and provide accurate responses.

Demo

Chat demo