#3535th of 54 in Agent frameworks and SDKsFor voice, speech and music
ten-framework
Framework for building real-time voice and multimodal conversational agents
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 11.2k
- License
- custom license
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Python
10,230 on 31 Mar 202611,184 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
TEN is a framework for real-time multimodal conversational agents, built from extensions for STT, LLM and TTS that are wired together in a graph. The repo ships example agents: a voice assistant over RTC or WebSocket, speech-to-speech, speaker diarization, lip-sync avatars, SIP calls, transcription and an ESP32-S3 board demo. A TMAN Designer UI on port 49483 edits extension properties.
Who it is for: Developers building low-latency voice agents with RTC, SIP or hardware clients
Strengths
- Many working voice examples: SIP calls, diarization, Live2D avatars, ESP32-S3 hardware
- Companion projects for VAD and turn detection, plus a memory extension example
- TMAN Designer lets you set STT, LLM and TTS keys in a graph UI
- Frontend and backend can be split, e.g. UI on Vercel or Netlify, backend on port 8080
Weaknesses
- License is Apache 2.0 with additional restrictions, not plain Apache 2.0
- Quick start needs Agora credentials plus OpenAI, Deepgram and ElevenLabs keys
- Quick start builds in a dev container and takes about 5-8 minutes
- Examples are mostly tied to hosted APIs; local-model setup is not documented in the README
What it needs
- RAM ≥ 4 GB
- no GPU
- Needs Docker, Node.js 18, Agora App ID and Certificate, OpenAI API key, Deepgram API key, ElevenLabs API key
- Models: OpenAI, Deepgram, ElevenLabs
- port 3000
Also in Agent frameworks and SDKs
See all 54| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 94 out of 100 |
| 2 | langchainPython framework for building LLM agents and apps from interoperable components | 89 out of 100 |
| 3 | browser-usePython library that lets an LLM agent control a web browser | 89 out of 100 |
| #4 | tradingagentsLangGraph multi-agent framework simulating a trading firm to analyze tickers | 88 out of 100 |
| #5 | copilotkitSDK for adding agent chat, generative UI and shared state to apps | 87 out of 100 |