Dograh

Freemium Code & Data

Dograh is an open-source platform for building voice agents that you can host on your own servers. It functions as a customizable alternative to proprietary solutions, allowing you to choose and combine your own AI models, speech processing, and communication channels.

Visual workflow builder that lets you drag and drop STT, LLM, TTS, and telephony components
Support for multiple model providers (OpenAI, Gemini, Groq, Azure, AWS, Hugging Face and others)
Self-hosted deployment on your own infrastructure, VPC, or fully air-gapped without external connectivity
Integration with Model Context Protocol to build agents directly from development environments like Claude Code and Cursor

✓ Pros

Complete data ownership: calls, transcripts, and customer information remain within your security boundary
Maximum flexibility to select and swap AI components, ensuring you're not locked into a single provider
Built-in compliance: supports HIPAA, GDPR, SOC 2, ISO 27001, and CCPA without depending on third-party audits

✕ Cons

Requires technical expertise to set up and maintain self-hosted deployments
Learning curve can be steeper compared to more simplified and closed-source solutions

Use cases

Build automated customer service systems that meet strict regulatory requirements in the healthcare sector

Develop outbound sales agents that make calls while maintaining maximum customer data privacy

Create custom voice systems for enterprises that need to keep sensitive information within their infrastructure

Visit Dograh →