huggingface/speech-to-speech

pBuild local voice agents with open-source models/phrdiv align=center
div
nbsp;
/div
img src=https://raw.githubusercontent.com/huggingface/speech-to-speech/main/logo.png width=600 /
h1Speech To Speech: Build voice agents with open-source models/h1
pa href=https://pypi.org/project/speech-to-speech/img src=https://img.shields.io/pypi/v/speech-to-speech alt=PyPI //a a href=https://pypi.org/project/speech-to-speech/img src=https://img.shields.io/pypi/pyversions/speech-to-speech alt=Python //a a href=https://raw.githubusercontent.com/huggingface/speech-to-speech/main/LICENSEimg src=https://img.shields.io/badge/license-Apache%202.0-blue alt=License //a/p
/div
pA low-latency, fully modular voice-agent pipeline: strongVAD -gt; STT -gt; LLM -gt; TTS/strong, exposed through an strongOpenAI Realtime-compatible WebSocket API/strong. Every component is swappable. The LLM slot speaks OpenAI-compatible protocols, so you can point it at a hosted provider, at a href=https://huggingface.co/inference-providersHF Inference Providers/a, or at a vLLM or llama.cpp server on your own hardware for a fully local, fully open stack./p
pThis pipeline runs in production as the conversation backend for thousands of a href=https://huggingface.co/blog/reachy-miniReachy Mini/a robots./p
p align=center
picture
source media=(prefers-color-scheme: dark) srcset=./docs/assets/endpoint-swap-dark.gif /
source media=(prefers-color-scheme: light) srcset=./docs/assets/endpoint-swap-light.gif /
img src=https://raw.githubusercontent.com/huggingface/speech-to-speech/main/docs/assets/endpoint-swap-light.gif alt=Switching an OpenAI Realtime client endpoint from hosted OpenAI to a self-hosted speech-to-speech server width=640 /
/picture /p
h2Quickstart/h2
precode class=language-bashpip install speech-to-speech
export OPENAI_API_KEY=...
speech-to-speech
/code/pre
pThis starts an OpenAI Realtime-compatible server at codews://localhost:8765/v1/realtime/code using Parakeet TDT for…