Dentro – AI Development & AI Consulting

AI Chatbot for Telegram

Ever wanted to use ChatGPT from the comfort of your messaging app? Now’s your chance! The Dentro AI bot allows you to send text and audio messages and receive responses on the go.

Telegram bot


Speech recognition

Custom instructions

What got us started

Two millionaires want to find out who’s richer, without telling each other how much they each have got in the bank.

This is an example for Yao’s millionaire problem, or a ’secure multi-party computational problem‘, to put it more formal.

We attempt to solve this with the help of artificial intelligence.

Starting a conversation is easy

The bot offers a quick introduction on how it can be used and then the start of a conversation is only a click away. 

Supporting both text and voice messages

Especially when on the go, voice messages sometimes are the easier way to operate a chat. The Dentro AI Bot understands them and translates them to text in order for the underlying AI to comprehend correctly what needs to be done. 

Fair usage limit with a frictionless upgrade option

Usage of the free version is limited to a generous limit of 10,000 tokens per user and day. This will generally be sufficient of most people. 

However, those who want to use more extensively have the option to enter their own OpenAI API key and enjoy unlimited usage. 

Making good use of built-in UI

Telegram offers several possibilities to enhance the user’s experience. The Dentro AI Bot utilizes buttons, input fields, links and static informational text in order to make the all processes as intuitive and fluid as possible. 

The technical side

We utilized the Python-Telegram-Bot library, one of the two major Python libraries, to form the backbone of our bot. GPT-4 by OpenAI serves as the brain behind the bot, enabling it to understand and respond to text in a way that feels almost human.

In order for the bot to be able to also ‚hear‘ you, we integrated Whisper by OpenAI, an incredible technology that transcribes voice messages into text. 

To ensure smooth deployment and make our bot accessible to everyone, we used Docker. 

The process behind the bot is as fascinating as the technology itself. When you send a voice message, Whisper transcribes it into text. That text is then fed to GPT-4, which crafts a response as if it were a human. No extra prompts, no fuss, just pure, intelligent conversation.