Open Source Voice Agent Platform

Dograh AI is an open-source, self-hostable platform designed to rapidly build and deploy voice AI agents. It serves as a FOSS alternative to proprietary solutions like Vapi, offering a drag-and-drop workflow builder for developers and power users. The platform provides full control, transparency, and aims for zero vendor lock-in, enabling users to create functional voice bots in under two minutes.

How It Works

Dograh AI employs a modular architecture and a visual drag-and-drop interface for workflow construction. It facilitates flexible integration with various Large Language Models (LLMs), Text-to-Speech (TTS), and Speech-to-Text (STT) services, allowing users to leverage Dograh's default stack or connect their own API keys. Key features include built-in telephony integration (e.g., Twilio) and AI testing personas called LoopTalk, which simulate real customer interactions for robust testing. This approach prioritizes speed, customization, and user control.

Quick Start & Requirements

• Primary install/run command:

  curl -o docker-compose.yaml https://raw.githubusercontent.com/dograh-hq/dograh/main/docker-compose.yaml && REGISTRY=ghcr.io/dograh-hq ENABLE_TELEMETRY=true docker compose up --pull always
  

• Prerequisites: Docker and Docker Compose.
• Resource footprint: Initial startup may take 2-3 minutes to download images. Access the dashboard at http://localhost:3010.
• Links:
  • Documentation: https://docs.dograh.com
  • Cloud Version: https://www.dograh.com

Highlighted Details

• Open-source alternative to Vapi with a stated 2-minute setup time.
• Self-hostable platform ensuring no vendor lock-in.
• Features LoopTalk (Beta) for AI-driven testing personas that mimic real customer behavior.
• Supports bringing your own LLM, TTS, STT, and telephony provider keys.
• Offers zero-config start with auto-generated Dograh API keys.

Maintenance & Community

The project is maintained by YC alumni and exit founders. Community engagement and support are primarily channeled through their Slack community, which is described as the cornerstone for contributions and discussions. Users can report bugs or request features via GitHub Issues.

Licensing & Compatibility

Dograh AI is licensed under the permissive BSD 2-Clause License. This license allows for free use, modification, and distribution, making it compatible with commercial applications and closed-source projects without imposing copyleft restrictions.

Limitations & Caveats

The LoopTalk AI testing feature is currently in Beta. Instructions for production self-hosted deployment are noted as "coming soon." By default, the platform collects anonymous usage data, which can be opted out by setting ENABLE_TELEMETRY to false.