Build a Telegram AI bot that handles voice and text with OpenAI

See What's Earning in AI Automation Freelancing.
DigiNo helps new AI automation freelancers earn faster by tracking what clients actually pay for.

Online coaches and content creators are losing potential clients to slow response times, but they cannot afford to hire support staff. This AI automation gives you a sellable, ready-to-deliver solution that answers voice and text messages around the clock inside Telegram.

What This Automation Does

Receives both typed messages and voice notes from Telegram users and routes each to the correct processing path
Transcribes voice notes into text using OpenAI so the AI agent can read and respond to spoken queries
Displays a live typing indicator in Telegram while the AI generates its response, making the bot feel human and responsive
Sends context-aware, conversational replies back to the user inside Telegram with memory of the ongoing conversation

Tools Used

Telegram
OpenAI
n8n

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 101 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 26 May 2026

Start Earning as a Freelancer on Contra

Contra is a commission-free professional network for independents. Browse live AI automation work and keep what you earn.

Join Contra Free →

How To Build It

Connect a Telegram bot to receive messages

Register a new bot through Telegram's BotFather to get an API token, then wire that token into the workflow so every incoming text or voice message from users triggers the automation instantly.

Route voice and text down separate paths

Add a conditional branch that detects whether the incoming message is a voice file or plain text, directing each to the correct handling path before any processing begins.

Transcribe voice notes into readable text

Pass any incoming voice file through OpenAI's transcription service so it is converted to clean text before reaching the AI agent, ensuring the bot can respond accurately to spoken input.

Configure the AI agent with conversation memory

Attach a windowed memory buffer to the AI agent so it retains context across multiple messages in a session, enabling the bot to hold a coherent multi-turn conversation rather than treating every reply as a fresh prompt.

Send the typing indicator and final reply to the user

Trigger a typing action in Telegram the moment processing starts, then deliver the AI-generated response back to the same chat once it is ready, giving users a smooth and natural interaction experience.

Pitfalls

Telegram webhooks silently stop delivering messages if your n8n instance loses its public URL or SSL certificate, so the bot goes dark with no error visible to the end user or the client.
OpenAI's transcription service struggles with heavy accents, background noise, or non-English voice notes, which can produce garbled text that causes the AI agent to reply with irrelevant or confusing answers.
The windowed memory buffer only retains a fixed number of recent messages, so long coaching conversations will lose early context and the bot may contradict or forget information the user already provided.

FAQ

Can I build this without coding?

Yes. The entire workflow is assembled through a visual interface using pre-built integrations for Telegram and OpenAI. You will need to generate API credentials and configure a webhook URL, but no custom code is required to deliver a working bot.

How long does it take?

A first build typically takes 90 minutes to two hours from a standing start, including bot registration and credential setup. Repeat builds for additional clients drop to under an hour once you have a tested template to clone.

What can I charge?

Pricing is yours to set based on the complexity of the system prompt, the number of revisions, and any ongoing maintenance you include. Clients in the coaching and creator space typically value the 24/7 availability and voice support as premium features worth scoping accordingly.

Which tool is required vs optional?

Telegram and OpenAI are both required for the core functionality. The voice transcription path depends on OpenAI's audio API, so if your client only needs text-based chat you can remove that branch entirely. The AI model itself can be swapped to Claude or another provider if a client has an existing preference.

This is original DigiNo analysis. The underlying automation pattern is a community workflow template – view the original on n8n.