An open-source WhatsApp bot that transcribes voice messages to text using local AI. Fully self-hosted, private, and blazing fast with GPU acceleration.
TekstoBot brings powerful speech-to-text capabilities to WhatsApp, running entirely on your own infrastructure.
Receive voice messages and instantly get accurate text transcriptions powered by Whisper AI — all processed locally.
Manage authorized numbers, view transcription history, and monitor connection status through an intuitive web interface.
All processing happens on your own server. No data leaves your infrastructure — your voice messages stay yours.
Automatic NVIDIA GPU detection and acceleration through CDI mapping. Falls back gracefully to CPU when no GPU is available.
Only pre-authorized phone numbers in your database can interact with the bot, giving you full control over access.
Built with Go's goroutines and worker pools, ensuring the bot stays responsive even under heavy load.
From voice message to readable text in seconds.
An authorized user sends a voice message via WhatsApp to the bot number.
Whisper AI processes the audio locally and converts speech to text with high accuracy.
The transcribed text is sent back to the user as a WhatsApp message, instantly.
A carefully chosen stack for performance, reliability, and developer experience.
Choose the installation method that best fits your environment.
# Install the TekstoBot RPM repository sudo dnf localinstall -y https://rpm.rda.run/repo.rpm
# Install TekstoBot from the repository sudo dnf install -y tekstobot
# Copy and edit the environment file sudo cp /etc/tekstobot.env.example /etc/tekstobot.env sudo vi /etc/tekstobot.env
# Enable and start TekstoBot (Whisper starts automatically) sudo systemctl enable --now tekstobot # Follow the logs sudo journalctl -u tekstobot -f
# Clone the repository git clone https://github.com/rda-run/tekstobot.git cd tekstobot # Configure environment variables cp .env.example .env # Edit .env with your database credentials # Install Go dependencies go mod tidy
# Run database migrations make migrate-up # Start the Whisper AI container make whisper
# Start the bot make run # Access the dashboard at http://localhost:8080
# Add the official NVIDIA repository curl -s -L https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo | \ sudo tee /etc/yum.repos.d/nvidia-container-toolkit.repo # Install the toolkit sudo dnf install -y nvidia-container-toolkit # Generate CDI hardware descriptors sudo mkdir -p /etc/cdi sudo nvidia-ctk cdi generate --output=/etc/cdi/nvidia.yaml