Dograh
Open-source voice platform for custom, self-hosted voice agents and pipelines.
About Dograh
Dograh is an open-source, self-hostable voice agent platform designed to give users full control over their voice data and workflows. It supports deploying on your own servers or cloud, with flexible model integration and real-time audio processing.
Features
- Modular workflow builder with plug-and-play components for STT, TTS, and telephony
- Fully open-source, self-hosted option for data sovereignty
- Supports custom models and real-time speech-to-speech interactions
- Multiple deployment options: self-host, managed cloud, or VPC deployment
- Compatibility with various models like Whisper, Voxtral, Kokoro, and custom voice cloning
- Designed for regulated industries with on-premise deployment compatibility
Who Is It For?
- Organizations requiring data sovereignty and compliance (fintech, healthtech, defense)
- Teams looking for customizable and forkable voice infrastructure
- Developers seeking flexible, open-source voice agent solutions
Use Cases
- Building AI call center and customer support agents
- Real-time translation and speech interaction in multiple languages
- Voice cloning and personalized voice assistant development
- On-premise deployment for sensitive data environments
Explore Dograh's open-source voice infrastructure for customizable, secure voice agents.
Open source
Dograh is an open-source voice platform that enables the creation of custom, self-hosted voice agents and pipelines. It offers a modular workflow builder for speech-to-text (STT), text-to-speech (TTS), and telephony components, supporting real-time audio processing and integration with various models. Designed for organizations in regulated industries, Dograh ensures data sovereignty and compliance while providing flexible deployment options. It caters to developers and teams seeking customizable voice infrastructure for applications such as AI call centers, real-time translation, and personalized voice assistants.