AI dictation in Telegram

Dictate a message — get neat text, ready to send.

AI assistants & Telegram bots · Voice & meetings

About the project

AI dictation turns voice messages into edited text right in Telegram. It is handy when typing by hand is inconvenient and you do not want to send the other person audio.

You dictate a message, and the system recognizes the speech, removes filler words and repetitions, and fixes spelling and style. The processing aims to preserve the original facts and meaning: a short remark stays short, while a long dictation becomes more readable.

Your own style is kept by default. If you need a formal, friendly or humorous tone, it is enough to ask. The commands “bold”, “new line” and “as a list” are used for formatting and do not end up in the finished text.

How it works

  1. Voice message in Telegram
  2. Speech recognition
  3. AI editing and formatting
  4. Finished text back in Telegram

Features

  • Voice input without switching between apps.

  • Cleaning the text of repetitions and filler words.

  • Correcting spelling and improving wording.

  • Keeping the author's style or changing the tone on request.

  • Controlling formatting with voice commands.

Where it can be used

  • Personal and business correspondence — when it is easier to say than to type.
  • Field staff — for preparing text messages to colleagues.
  • Communication with customers — for turning a spoken wording into a neat reply.
  • Preparing notes and drafts — the approach can be adapted for capturing ideas by voice.

Technologies

  • n8n
  • Telegram
  • Whisper
  • GPT-4.1

More projects

All projects

AI assistants & Telegram bots

Calendar AI & Chronos

Plan your tasks and log working time — with an ordinary message in Telegram.

A personal AI assistant: Google Calendar events and time tracking in Google Sheets from a text or voice message.

Result Create or view a calendar event in 10–15 seconds

  • Telegram
  • n8n
  • LLM
  • Speech recognition
  • Google Calendar
  • +1

Voice & meetings · Documents

Local transcription of large audio files

Hours of recordings turned into ready text — without cloud service limits.

Batch transcription of audio recordings on your own computer with GPU acceleration, fully offline.

Result About 60 hours of lectures processed in about 5 hours on a GPU

  • Python
  • Faster Whisper
  • NVIDIA CUDA
  • FFmpeg
  • n8n
  • +2

Voice & meetings · Test automation & QA

Voice bug report

Tell Telegram about a bug — get a formatted report and a Jira task.

A tester's voice notes turn into structured bug reports in English.

Result Generate a bug report in 60 seconds

  • Telegram
  • n8n
  • OpenAI
  • Google Sheets
  • Google Docs
  • +1

Start with one process.

Describe what is currently done manually, what data is involved and what result you would like to achieve. You do not need a technical specification or a chosen model to begin.

Fill in the brief
AI bot consultation (opens Telegram, an external service)

Please do not share trade secrets, source code, or sensitive personal data before an NDA (non-disclosure agreement) is signed.