Vexa作为一个开源项目,支持本地部署,适合具备技术能力的企业或个人用户。以下是详细的部署步骤和硬件要求:
Deployment process
- Cloning Warehouse:
git clone https://github.com/Vexa-ai/vexa.git cd vexa
- 初始化子模块:
make submodules
- Configure environment variables:
make env
编辑.env文件设置参数(如ADMIN_API_TOKEN等)。
- Download the Whisper model:
make download-model
- Build the Docker image:
docker build -t vexa-bot:latest -f services/vexa-bot/core/Dockerfile ./services/vexa-bot/core
- Start the service:
docker compose build docker compose up -d
hardware requirement
- Recommended Configurations:配备NVIDIA GPU的服务器
- minimum requirement:16GB RAM + 4核CPU
- storage space:需预留足够空间存放Whisper模型(默认存储在./hub目录)
服务启动后,API网关运行在http://localhost:8056,管理接口位于http://localhost:8057。需定期执行git pull和docker compose up –build以获取最新功能。
This answer comes from the articleVexa: a real-time meeting transcription and intelligent knowledge extraction toolThe