🔧 What We'll Build
A Telegram chatbot that:
Runs on Python
Listens to user messages on Telegram
Sends the messages to OpenAI’s GPT API
Replies with a GPT-generated response
✅ Prerequisites
1. Python 3.7+
2. Install Python packages:
pip install python-telegram-bot openai
3. Get:
o A Telegram Bot Token from BotFather
o An OpenAI API Key from https://platform.openai.com
🧠 Step-by-Step Tutorial
Step 1: Import Libraries
from telegram import Update
from telegram.ext import ApplicationBuilder, ContextTypes, CommandHandler,
MessageHandler, filters
import openai
import os
Step 2: Set Your API Keys
openai.api_key = "YOUR_OPENAI_API_KEY"
TELEGRAM_TOKEN = "YOUR_TELEGRAM_BOT_TOKEN"
💡 For security, you should ideally load these from environment variables using
os.getenv("KEY_NAME").
Step 3: Create the GPT Chat Function
async def chat_with_gpt(message: str) -> str:
try:
response = openai.ChatCompletion.create(
model="gpt-4",
messages=[{"role": "user", "content": message}],
)
return response['choices'][0]['message']['content'].strip()
except Exception as e:
return f"Error: {str(e)}"
Step 4: Define Telegram Handlers
async def start(update: Update, context: ContextTypes.DEFAULT_TYPE):
await update.message.reply_text("Hi! I'm your AI chatbot. Just type
anything to chat!")
async def handle_message(update: Update, context: ContextTypes.DEFAULT_TYPE):
user_message = update.message.text
reply = await chat_with_gpt(user_message)
await update.message.reply_text(reply)
Step 5: Run the Bot
def main():
app = ApplicationBuilder().token(TELEGRAM_TOKEN).build()
app.add_handler(CommandHandler("start", start))
app.add_handler(MessageHandler(filters.TEXT & ~filters.COMMAND,
handle_message))
print("Bot is running...")
app.run_polling()
if __name__ == "__main__":
main()
✅ Summary
Feature Description
Platform Telegram
Intelligence OpenAI GPT (any model: GPT-3.5/4)
Framework python-telegram-bot
Language Python
Mode Polling (easy to deploy anywhere)
🧠 What is ChatterBot?
ChatterBot is a Python library that uses machine learning to generate responses based on past
conversations. It can be trained using preloaded datasets or your own data.
✅ Prerequisites
1. Install ChatterBot and dependencies:
bash
CopyEdit
pip install chatterbot==1.0.5
pip install chatterbot_corpus
⚠️chatterbot development has slowed down, so we use version 1.0.5 which is stable with
Python 3.7–3.9.
🧪 Step-by-Step Example: ChatterBot in Action
Step 1: Basic Bot Setup
from chatterbot import ChatBot
from chatterbot.trainers import ChatterBotCorpusTrainer
# Create chatbot instance
chatbot = ChatBot('SimpleBot')
# Create and train the bot using English corpus
trainer = ChatterBotCorpusTrainer(chatbot)
trainer.train("chatterbot.corpus.english")
Step 2: Interactive Chat Loop
print("SimpleBot is ready to chat! Type 'exit' to stop.")
while True:
user_input = input("You: ")
if user_input.lower() == 'exit':
break
response = chatbot.get_response(user_input)
print("Bot:", response)
🧠 Sample Output
txt
You: Hello
Bot: Hi there!
You: What is your name?
Bot: My name is SimpleBot.
You: exit
🔁 How to Improve It
Custom training:
from chatterbot.trainers import ListTrainer
trainer = ListTrainer(chatbot)
trainer.train([
"Hi there!",
"Hello!",
"How are you?",
"I'm good, thank you.",
])
Multi-language support: train with "chatterbot.corpus.french" or other languages.
Save/load conversations: ChatterBot uses a SQLite database (db.sqlite3 by default).
⚠️Limitations
Doesn't use modern NLP (like transformers or GPT).
Basic conversation only — it doesn’t understand context or intent deeply.
✅ Summary
Feature Description
Library chatterbot
Intelligence Learns from corpus or user input
Training Prebuilt or custom conversations
Good for Learning, demos, offline use
