wx44wx/py-active-call

Go to file

Xin Wang 294a3e405c sentences ends update

2026-02-06 07:55:06 +08:00

Init Projecto

2026-01-28 10:19:04 +08:00

Add heartbeat

2026-02-04 23:16:30 +08:00

sentences ends update

2026-02-06 07:55:06 +08:00

update client latency and three utterances example

2026-02-06 07:51:09 +08:00

fix long run bug

2026-02-03 12:05:09 +08:00

update client latency and three utterances example

2026-02-06 07:51:09 +08:00

Add heartbeat

2026-02-04 23:16:30 +08:00

Fix Duplicate / inconsistent EOU

2026-02-06 07:23:31 +08:00

api has llm response event

2026-02-04 12:00:52 +08:00

a basic duplex agent is built

2026-01-29 16:36:46 +08:00

Integrate eou and vad

2026-01-29 13:57:12 +08:00

.env.example

I can use text to get audio response and barge in

2026-01-29 16:25:53 +08:00

.gitignore

update gitignore

2026-02-04 14:16:27 +08:00

pyproject.toml

Integrate eou and vad

2026-01-29 13:57:12 +08:00

README.md

fix long run bug

2026-02-03 12:05:09 +08:00

requirements.txt

I can use text to get audio response and barge in

2026-01-29 16:25:53 +08:00

README.md

py-active-call-cc

Python Active-Call: real-time audio streaming with WebSocket and WebRTC.

This repo contains a Python 3.11+ codebase for building low-latency voice pipelines (capture, stream, and process audio) using WebRTC and WebSockets. It is currently in an early, experimental stage.

Usage

启动

uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

测试

python examples/test_websocket.py

python mic_client.py