# Voice Agents

> AI agents that hold real-time voice conversations for calls, support, and assistants.

Voice agents hold real-time spoken conversations — answering and placing phone calls, running support lines, and powering assistants — by stitching speech-to-text, an LLM, and text-to-speech together under tight latency budgets. The differentiators are conversational latency and naturalness, telephony and SIP integration, and how gracefully the agent handles the hard parts of real calls: interruptions, silence, background noise, and a clean hand-off to a human when it's out of depth. Sub-second response time is what separates a natural conversation from an awkward one, so scrutinize end-to-end latency, not just the quality of the underlying model. Evaluate the available voices, how well the system handles barge-in when a caller talks over it, and how it manages turn-taking and context across a long call. Integration surface matters too: check telephony providers, warm transfer to human agents, function calling into your back-end systems, and analytics on call outcomes. Most platforms price per minute of conversation, bundling speech, model, and telephony costs, sometimes with a monthly platform fee on top — so model your real call volume before comparing headline rates. Test with your own scripts, accents, and edge cases rather than the vendor's demo, because voice quality degrades fastest exactly where real callers live: noisy lines, fast talkers, and unexpected questions.

29 tools indexed. Sponsored placements are labeled and never ranked; open-source ordering reflects real GitHub adoption. Browse on the web: https://agentsearchengine.app/category/voice-agents

## Commercial products (alphabetical — stars don't measure commercial adoption)
- [Bland AI](https://agentsearchengine.app/agents/bland-ai) — Enterprise platform for AI phone calls that sound human. · Paid product · Platform · best for enterprises running high-volume outbound and inbound ai phone calls
- [ElevenLabs Agents](https://agentsearchengine.app/agents/elevenlabs-agents) — Conversational voice agents powered by ElevenLabs' leading speech models. · Freemium · Platform · best for voice agents that need best-in-class speech quality and voice cloning
- [Retell AI](https://agentsearchengine.app/agents/retell-ai) — Build production-ready AI phone agents with ultra-low latency. · Paid product · Platform · best for production ai phone agents with low latency and strong call-handling primitives
- [Synthflow](https://agentsearchengine.app/agents/synthflow) — No-code AI voice agents for inbound and outbound calls. · Paid product · Platform · best for no-code teams launching ai phone agents quickly
- [Vapi](https://agentsearchengine.app/agents/vapi) — Developer platform for building, testing, and deploying voice AI agents. · Paid product · Platform · best for developers building production voice agents with full api control over models, voices, and telephony

## Open-source index (ranked by adoption)
1. [xiaozhi-esp32-server](https://agentsearchengine.app/agents/xiaozhi-esp32-server) — 本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server. · Open source · Infrastructure · 10k★ · JavaScript · MIT · active 5 days ago
2. [ten-vad](https://agentsearchengine.app/agents/ten-vad) — Voice Activity Detector (VAD): low-latency, high-performance and lightweight · Open source · SDK / library · 2.2k★ · C · active 5 months ago
3. [bailing](https://agentsearchengine.app/agents/bailing) — 百聆 是一个类似GPT-4o的语音对话机器人，通过ASR+LLM+TTS实现，集成DeepSeek R1等优秀大模型，接入openClaw，真正的个人语音助手，时延低至800ms，Mac等低配置也可运行，支持打断 · Open source · Agent · 1.7k★ · Python · MIT · active 3 months ago · best for experimenting with a low-latency chinese-language voice assistant on modest hardware
4. [RCLI](https://agentsearchengine.app/agents/rcli) — Talk to your Mac, query your docs, no cloud required. On-device voice AI + RAG · Open source · Agent · 1.5k★ · C++ · MIT · active 4 months ago
5. [CyberVerse](https://agentsearchengine.app/agents/cyberverse) — Self hosted, real-time digital human agent platform. Build voice-first AI agents with WebRTC, persona memory, tools, RAG, and optional digital-human video. · Open source · Platform · 1.5k★ · Python · GPL-3.0 · active 4 days ago
6. [Patter](https://agentsearchengine.app/agents/patter) — Open-source voice-AI SDK. The Vapi/Retell alternative for builders who want to own the stack. · Open source · SDK / library · 1.0k★ · Python · MIT · active 1 day ago
7. [rustpbx](https://agentsearchengine.app/agents/rustpbx) — A PBX written by rust · Open source · Infrastructure · 736★ · Rust · MIT · active today
8. [echokit_server](https://agentsearchengine.app/agents/echokit-server) — Open Source Voice Agent Platform · Open source · Platform · 589★ · Rust · GPL-3.0 · active 5 months ago
9. [Starmoon](https://agentsearchengine.app/agents/starmoon) — A conversational, AI device + software framework for companionship, entertainment, education, healthcare, IoT applications, and DIY robotics. · Open source · Framework · 549★ · TypeScript · GPL-3.0 · active 4 months ago
10. [JARVIS](https://agentsearchengine.app/agents/jarvis) — Your own personal voice assistant: Voice to Text to LLM to Speech, displayed in a web interface · Open source · Agent · 526★ · Python · GPL-3.0 · active 2 years ago
11. [deepgram-python-sdk](https://agentsearchengine.app/agents/deepgram-python-sdk) — Official Python SDK for Deepgram. · Open source · SDK / library · 454★ · Python · MIT · active 3 days ago
12. [bolna](https://agentsearchengine.app/agents/bolna) — End-to-end platform for building voice first multimodal agents · Open source · Agent · 432★ · Python · MIT · active 1 year ago
13. [sutando](https://agentsearchengine.app/agents/sutando) — My AI Stand. Realtime by day, rewriting itself by night. Summon my AI superpower. · Open source · Agent · 359★ · Python · MIT · active today
14. [ai-avatar-system](https://agentsearchengine.app/agents/ai-avatar-system) — AI Avatar / digital human platform — upload a photo, clone a voice, talk to any face in real time with lip-sync video. Open-source, self-hosted. Claude · Whisper · Chatterbox · MuseTalk. · Open source · Agent · 344★ · Python · MIT · active 5 days ago
15. [agent-second-brain](https://agentsearchengine.app/agents/agent-second-brain) — An always-on second brain you talk to. Voice notes in Telegram → typed, linked knowledge in your Obsidian vault. Runs 24/7 on the Claude subscription you already have. · Open source · Agent · 342★ · Python · MIT · active 18 days ago
16. [xybrid](https://agentsearchengine.app/agents/xybrid) — Build apps powered by on-device AI · Open source · SDK / library · 319★ · Rust · Apache-2.0 · active 1 day ago
17. [LLMVoX](https://agentsearchengine.app/agents/llmvox) — LLMVoX: Autoregressive Streaming Text-to-Speech Model for Any LLM · Open source · SDK / library · 308★ · Python · active 1 year ago
18. [deepgram-js-sdk](https://agentsearchengine.app/agents/deepgram-js-sdk) — Official JavaScript SDK for Deepgram. · Open source · SDK / library · 269★ · TypeScript · MIT · active 3 days ago
19. [M.I.L.E.S](https://agentsearchengine.app/agents/m-i-l-e-s) — M.I.L.E.S, a GPT-4-Turbo voice assistant, self-adapts its prompts and AI model, can play any Spotify song, adjusts system and Spotify volume, performs calculations, browses the web and internet, · Open source · Agent · 233★ · Python · MIT · active 8 days ago
20. [gdansk-ai](https://agentsearchengine.app/agents/gdansk-ai) — Full stack voice chatbot · Open source · Agent · 199★ · TypeScript · GPL-2.0 · active 1 year ago
21. [active-call](https://agentsearchengine.app/agents/active-call) — A SIP/WebRTC voice agent · Open source · Agent · 184★ · Rust · active 3 days ago
22. [axiom-voice-agent](https://agentsearchengine.app/agents/axiom-voice-agent) — Run a <400ms latency Voice Agent on just 4GB VRAM. Fully offline, no API keys required. Optimized for GTX 1650 and edge robotics with zero-copy inference. (Apache 2.0) · Open source · Agent · 136★ · Python · Apache-2.0 · active 2 months ago
23. [BentoVoiceAgent](https://agentsearchengine.app/agents/bentovoiceagent) — Build Phone Calling Voice Agent fully powered by open source models. · Open source · Agent · 125★ · Python · active 20 days ago
24. [AI-Voice-Agent](https://agentsearchengine.app/agents/ai-voice-agent) — Self-hosted AI voice agent · Open source · Agent · 35★ · Python · MIT · active 11 days ago

## Frequently asked
**What is a voice AI agent?**
A voice agent conducts natural, real-time voice conversations — answering and making calls, handling support, or acting as an assistant — by combining speech recognition, an LLM, and voice synthesis.

**How do I choose a voice agent platform?**
Prioritize conversational latency and naturalness, telephony/SIP integration, interruption handling, and how easily it hands off to a human. Test with your real call scripts.

**How much do voice agents cost?**
Typically per-minute usage pricing that bundles speech, model, and telephony costs; some platforms add a monthly fee.

---
Source: [Agent Search Engine](https://agentsearchengine.app). Full directory in one file: https://agentsearchengine.app/llms-full.txt