hkjarral/AVA-AI-Voice-Agent-for-Asterisk
An open-source AI Voice Agent that integrates with Asterisk/FreePBX using Audiosocket/RTP technology
What it solves
AVA is an open-source AI voice agent designed for Asterisk and FreePBX phone systems. It allows developers and enterprises to integrate sophisticated AI-driven voice interactions into their telephony infrastructure, replacing or augmenting traditional IVRs with modular, flexible AI agents.
How it works
The project uses a modular pipeline architecture that separates the voice interaction process into three main components: Speech-to-Text (STT), Large Language Models (LLM), and Text-to-Speech (TTS). Users can mix and match different providers for each stage of the pipeline. It integrates with Asterisk via the Asterisk REST Interface (ARI) and provides an Admin UI for configuration, agent management, and call history tracking. It also supports outbound calling campaigns and integration with VICIdial.
Who it’s for
It is built for developers, system administrators, and enterprises using Asterisk or FreePBX who want to deploy production-ready AI voice agents for customer service, sales, or other automated phone interactions.
Highlights
- Modular Pipeline: Mix and match STT, LLM, and TTS providers.
- Enterprise Ready: Includes six production-validated "golden baselines" for deployment.
- Admin Dashboard: A web-based UI for managing agents, providers, and monitoring calls.
- Telephony Integration: Native support for Asterisk/FreePBX, including dialplan configuration and VICIdial integration.
- Advanced Audio Support: Supports HD Voice (16 kHz AudioSocket) and enhanced telephony audio for cleaner playback.
- Outbound Capabilities: Tools for importing leads via CSV/Excel and scheduling outbound AI-driven calls.
関連
- プロジェクト
- プロジェクト
- プロジェクト
- プロジェクト
- プロジェクト