经 AI Skill Hub 精选评估,快速语音 获评「推荐使用」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 7.5 分,适合有一定技术背景的用户使用。
快速语音 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
快速语音 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:npm 全局安装 npm install -g quickvoice # 方式二:npx 直接运行(无需安装) npx quickvoice --help # 方式三:项目依赖安装 npm install quickvoice # 方式四:从源码运行 git clone https://github.com/allgpt-co/QuickVoice cd QuickVoice npm install npm start
# 命令行使用
quickvoice --help
# 基本用法
quickvoice [options] <input>
# Node.js 代码中使用
const quickvoice = require('quickvoice');
const result = await quickvoice.run(options);
console.log(result);
# quickvoice 配置说明 # 查看配置选项 quickvoice --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export QUICKVOICE_CONFIG="/path/to/config.yml"
Open-source AI phone-agent infrastructure you can run, inspect, and extend.
QuickVoice is the open-source Retell alternative for engineering-led teams that want control over the voice-agent stack instead of only consuming a closed hosted API. It gives teams the full product surface in one repo: a marketing site, customer console, API server, LiveKit-powered AI worker, telephony integrations, knowledge bases, call logs, outbound campaigns, billing paths, MCP/tool connections, privacy controls, and local development tooling.
Website: quickvoice.co
Project status: QuickVoice is under active development and has not published a stable release yet. Review the roadmap, draft v0.1.0 notes, and known setup boundaries before evaluating it for production.
<p align="center"> <img src="./apps/web/public/dashboard.png" alt="QuickVoice console dashboard preview" width="920"> </p>
apps/web - Next.js website with product pages, use cases, industry pages, blog content, pricing, and legal pages.apps/console - Next.js customer console for organizations, agents, numbers, calls, knowledge bases, API keys, billing, and settings.apps/server - Express API server with auth, permissions, agent configuration, phone numbers, call logs, outbound calls, MCP connections, retention jobs, Stripe, Twilio, Telnyx, LiveKit, S3, Redis/BullMQ, and Inngest integrations.apps/ai - Python AI service and LiveKit worker handlers for runtime configuration, RAG, MCP tool execution, privacy controls, call logging, recording, and voice-agent execution.packages/eslint-config and packages/typescript-config - Shared monorepo linting and TypeScript configuration.scripts, Taskfile.yml, and docker-compose.dev.yml - Local orchestration for Node services, Python services, Prisma, Postgres, and Redis.The Docker Compose database credentials are dev-only placeholders (quickvoice / quickvoice) and the Postgres and Redis ports are bound to 127.0.0.1. Edit the generated env files after the first run if you need real Google, Stripe, LiveKit, Twilio, Telnyx, SMTP, or AWS credentials. Generated env files are ignored by git.
The placeholder values are enough to inspect local startup paths; they do not make provider-backed actions work. A successful task up:dev does not prove that live calls, OAuth, billing, email delivery, object storage, Pinecone retrieval, or production deployment are configured. Real phone calls require LiveKit plus a configured telephony provider and model-provider credentials. Keep production data and credentials out of local issue reproductions.
To exercise prepaid billing safely, follow the Stripe wallet test-mode verification guide. It keeps wallet and legacy subscription webhook secrets separate and verifies webhook-authoritative crediting, duplicate delivery, and refunds against a Stripe sandbox.
Optional local email testing is available through a Docker Compose profile:
docker compose -f docker-compose.dev.yml --env-file .env.dev --profile mail up -d mailpit
Useful individual tasks:
task doctor
task env:dev
task docker:up
task db:migrate
task db:seed -- --email you@example.com
task ci
task server:dev
task console:dev
task web:dev
task ai:api
task ai:worker
Common root commands:
pnpm build
pnpm lint
pnpm check-types
pnpm test
pnpm ci:local
pnpm audit:deps -- --audit-level low
The supported one-command development path is a Linux environment with:
>=4^20.19 || ^22.13 || >=24 and Corepackpnpm@9.0.0 (activated by the setup task)Then run:
task up:dev
task up and task dev are aliases. Go Task treats spaces as separate task names, so prefer task up:dev for the explicit form.
开源AI电话代理工作流,替代Retell,值得关注
该工具使用 AGPL-3.0 协议,商用场景请仔细阅读协议条款,必要时咨询法律意见。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
⚠️ AGPL 3.0 — 最严格的 Copyleft,网络服务端使用也需开源,SaaS 使用受限。
AI Skill Hub 点评:快速语音 的核心功能完整,质量良好。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | QuickVoice |
| 原始描述 | 开源AI工作流:The open-source alternative to Retell for AI phone agents.。⭐10 · TypeScript |
| Topics | ai-voice-agentscall-automationexpresslivekitnextjstypescript |
| GitHub | https://github.com/allgpt-co/QuickVoice |
| License | AGPL-3.0 |
| 语言 | TypeScript |
收录时间:2026-06-24 · 更新时间:2026-06-24 · License:AGPL-3.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端