LingEcho is an enterprise-grade intelligent voice interaction platform providing a comprehensive AI voice solution. It integrates advanced speech recognition (ASR), text-to-speech (TTS), large language models (LLM), and real-time communication technologies. The platform targets businesses seeking to implement sophisticated AI-driven voice capabilities, offering features like real-time calls, voice cloning, and workflow automation to enhance customer engagement and operational efficiency.

How It Works

The platform is architected with a Go backend (using Gin) and a React frontend. Core functionalities are powered by integrating specialized AI services: WebRTC for real-time calls, SileroVAD for voice activity detection, ModelScope for voiceprint recognition, Whisper for ASR, and edge-tts for TTS. It supports LLMs for intelligent responses and features a visual workflow designer for automating complex business processes, alongside a knowledge base management system for intelligent data retrieval.

Quick Start & Requirements

• Primary install: Docker Compose is the recommended method for setup (docker-compose up -d).
• Prerequisites: Go >= 1.24.0, Node.js >= 18.0.0, npm/pnpm >= 8.0.0, Git. Python >= 3.10 is required for optional services (VAD, Voiceprint, ASR-TTS). Docker & Docker Compose are recommended.
• Links: Live Demo: https://lingecho.com, API Documentation: http://localhost:7072/api/docs.

Highlighted Details

• AI Character Real-time Calls leveraging WebRTC for low-latency interaction.
• Custom Voice Cloning and Training capabilities for personalized AI assistant voices.
• Visual Workflow Automation designer supporting multiple trigger types for complex business process automation.
• Knowledge Base Management system with document storage, retrieval, and AI analysis features.
• Standalone ASR (Whisper) and TTS (edge-tts) services for flexible integration.

Maintenance & Community

The project is maintained by a core team of two full-stack engineers: chenting (Project Manager, Architecture) and wangyueran (Frontend Development). Direct contact is available via email at 19511899044@163.com. No community channels like Discord or Slack are explicitly mentioned in the README.

Licensing & Compatibility

The project is licensed under the MIT License. This license is permissive and generally allows for commercial use, modification, and distribution, including within closed-source applications.

Limitations & Caveats

The provided README does not explicitly detail limitations, alpha status, or known bugs. The setup for optional Python-based services (VAD, Voiceprint, ASR-TTS) requires separate configuration and management, potentially adding complexity to the deployment.