#7373rd of 88 in VoiceFor voice, speech and music
patter
SDK that connects an AI voice agent to phone calls via Twilio, Telnyx or Plivo
- Stars
- 1.1k
- License
- MIT
- Last commit
- Aug 2026
- Last release
- Jun 2026
- Language
- Python
4 on 21 Apr 20261,083 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Patter is a Python and TypeScript SDK that runs the agent loop for phone calls, covering LLM, speech-to-text, text-to-speech, realtime voice engines and the telephony carrier. Each layer is a swappable provider, and calls run in Realtime, Pipeline or Hybrid mode. A built-in Cloudflare tunnel and dashboard support local development, and calls can be simulated from the terminal.
Who it is for: Developers building voice agents that answer or place phone calls
Strengths
- Python and TypeScript SDKs with the same API, hooks and events
- Swappable providers per layer: 5 LLMs, 7 STT, 7 TTS, 4 realtime engines
- Works with Twilio, Telnyx and Plivo; tools, transfer and guardrails are carrier-independent
- LLM fallback chain mid-call and OpenTelemetry traces per call
Weaknesses
- Depends on paid cloud providers and a carrier account; no local model path is documented
- Anonymous telemetry is on by default and must be disabled by opt-out
- Quick tunnel is for development; production needs a static webhook URL or ngrok
- Templates live in separate repos, not in this one
What it needs
- no GPU
- Docker + Compose
- Needs Twilio, Telnyx or Plivo account, LLM/STT/TTS provider API keys, Cloudflare quick tunnel or ngrok
- Models: OpenAI, Anthropic, Google Gemini, Groq, Cerebras
What changed
- v0.7.0 ↗ (opens in a new tab)Security fix
“SSRF guard now folds non-canonical IP encodings — decimal/octal/hex/short IPv4 and IPv4-mapped IPv6 (incl. 169.254.169.254 metadata). Shared getpatter/ssrf.py.”
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Voice
See all 88| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 98 out of 100 |
| 2 | handyOffline push-to-talk dictation app that pastes transcripts into any text field | 91 out of 100 |
| 3 | whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally | 90 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 84 out of 100 |
| #5 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 83 out of 100 |