Agent Voice Response (commonly abbreviated AVR) is a conversational AI platform that integrates with the open‑source PBX Asterisk to replace traditional IVR (Interactive Voice Response) systems using AI-driven voice interactions. It acts as a voicebot, transcribing speech, generating responses through large language models (LLMs), and synthesizing speech back to callers.
AVR offers a Docker-based deployment enabling organizations to integrate automatic speech recognition (ASR), LLM, text-to-speech (TTS), and direct speech-to-speech (STS) in voice calling systems. It supports a flexible, provider-agnostic architecture. Users can plug in services such as Google, Deepgram for ASR/TTS, and OpenAI or Anthropic for LLM, or opt for full voice‑to‑voice pipelines like OpenAI Realtime or Ultravox.ai.
- Modular AI Components: Swappable ASR, LLM, and TTS Blocks via HTTP APIs
- Voice Activity Detection (VAD): Enables natural interruption in voice calls
- Dockerized Microservices: Each core service runs isolated in containers for scalability
- Support for Function‑Calling: Integration with OpenAI and Anthropic assistive features
- Speech‑to‑Speech Mode: Direct voice transformation using providers like OpenAI Realtime
- Multi‑language & Custom Voices: Offerings include configurable voice personalities and language support
- Core Service – Orchestrates voice input/output via Asterisk and AudioSocket
- ASR Service – Transcribes caller audio into text
- LLM Service – Generates conversational responses
- TTS Service – Converts responses back to speech
- Web GUI – (Under development) for deployment management
- Asterisk Integration – Communication via SIP and AudioSocket
The avr-infra GitHub repository provides Docker Compose templates integrating AVR Core, Asterisk, and configurable ASR/LLM/TTS stacks (e.g., Google, Deepgram, OpenAI, Anthropic, Vosk).
Examples include combinations like Deepgram + OpenAI, Google + OpenRouter, ElevenLabs + Anthropic, and full voice pipelines via OpenAI Realtime or Ultravox.
¶ Use and Community
- Installation: Requires Docker and Asterisk (v18+ with AudioSocket); users register a SIP client and call a defined extension to interact with the AI agent
- Licensing: Distributed free for personal and commercial use; the AVR source remains proprietary though the infra and integration repos are open‑source
- Contributions: Maintainers host a Discord community and accept donations via Ko‑fi
Some notable public repositories under the agentvoiceresponse organization include:
- avr-infra – Docker Compose templates for deploying AVR with Asterisk and AI services
- avr-llm-openai, avr-llm-openai-assistant – OpenAI-based LLM connectors
- avr-asr-google-cloud-speech, avr-asr-vosk – ASR modules leveraging Google Speech or open-source Vosk
- avr-tts-google-speech-tts, avr-tts-openai, avr-tts-coquitts – TTS connectors for various engines
- avr-sts-openai, avr-sts-ultravox - STS connectors for varous engines
- avr-vad – VAD library (Silero) for speech activity detection, available as an npm package
- avr-asterisk – Docker image of Asterisk optimized for AVR environments