Agent Voice Response (AVR) is an open-source infrastructure designed to connect modern conversational AI systems (LLMs, ASR, TTS) with traditional telephony platforms such as Asterisk and FreePBX.
The project makes it possible to:
- Integrate cutting-edge AI models (speech recognition, natural language, speech synthesis) into real-time phone calls.
- Provide a vendor-agnostic framework: you can choose your preferred ASR, TTS, or LLM provider (e.g., Deepgram, Google, OpenAI, ElevenLabs).
- Simplify deployment through Docker-based services, making experimentation and production setups much faster.
- Enable developers, researchers, and companies to innovate in voice automation without being locked to proprietary platforms.
Maintaining AVR requires:
- Continuous updates to stay compatible with rapidly evolving AI providers.
- Infrastructure for testing across different environments (Linux distros, cloud providers, PBX setups).
- Documentation, tutorials, and community support.
Your contribution helps ensure AVR remains free, open-source, and actively developed.
If you find AVR useful and want to see it grow, please consider supporting the project:
Your donation helps sustain development and encourages further improvements. Even a small contribution makes a big difference!
🙏 Thank you for being part of the community and helping to build the future of voice + AI integration!