Telegram • AI Audio

STT TG Bot

High-speed speech-to-text and voice generation Telegram bot powered by Groq Whisper and OpenAI TTS.

Overview

High-speed Telegram bot for bidirectional speech processing: instant transcription of audio into text (STT) and text-to-speech synthesis (TTS). Accepts voice messages, audio files, and video circles in any format, rapidly transcribing them via Groq Whisper. Also supports voice synthesis via OpenAI TTS, user whitelisting, and a production FastAPI webhook architecture.

Architecture & Tech Stack

Leverages low-latency inference endpoints to process audio in sub-second time, running as a stateless webhook service.

Bot & Webhooks: Python 3.11 · aiogram 3 · FastAPI
AI Models: Groq Whisper (whisper-large-v3-turbo) · OpenAI TTS
Deployment: Railway · Docker