Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
sub-500ms latency phone agent orchestration
| Date | Stars |
|---|---|
| 2026-07-31 | 668 |
| 2026-08-01 | 668 |
| 2026-08-06 | 668 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# shuo 说
A voice agent framework in ~600 lines of Python.
```bash
python main.py +1234567890
```
```
🚀 Server starting on port 3040
✓ Ready https://mature-spaniel-physically.ngrok-free.app
📞 Calling +1234567890...
✓ Call initiated SID: CA094f2e...
🔌 WebSocket connected
▶ Stream started SID: MZ8a3b1f...
← Flux EndOfTurn "Hey, how's it going?"
◆ LISTENING → RESPONDING
→ Start Agent "Hey, how's it going?"
← Agent turn done
◆ RESPONDING → LISTENING
```
## How it works
Two abstractions, one pure function:
- **Deepgram Flux** — always-on STT + turn detection over a single WebSocket
- **Agent** — self-contained LLM → TTS → Player pipeline, owns conversation history
- **`process_event(state, event) → (state, actions)`** — the entire state machine in ~30 lines
Everything streams. LLM tokens feed TTS immediately, TTS audio feeds Twilio immediately. If you interrupt (barge-in), the agent cancels everything and clears the audio buffer instantly.
```
LISTENING ──EndOfTurn──→ RESPONDING ──Done──→ LISTENING
↑ │
└────StartOfTurn─────────┘ (barge-in)
```
## Project structure
```
shuo/
types.py # Immutable state, events, actions
state.py # Pure state machine (~30 lines)
conversation.py # Main event loop
agent.py # LLM → TTS → Player pipeline
log.py # Colored logging
server.py # FastAPI endpoints
services/
flux.py # Deepgram Flux (STT + turns)
llm.py # OpenAI GPT-4o-mini streaming
tts.py # ElevenLabs WebSocket streaming
tts_pool.py # TTS connection pool (warm spares)
player.py # Audio playback to Twilio
twilio_client.py # Outbound calls + message parsing
```
## Setup
Requires Python 3.9+, [ngrok](https://ngrok.com/), and API keys for Twilio, Deepgram, OpenAI, and ElevenLabs.
Configure your [ngrok authentication token](https://dashboard.ngrok.com/get-started/your-authtoken):
```bash
ngrok config add-authtoken <YOUR_NGROK_AUTH_TOKEN>
```
```bash
pip install -r requirements.txt
cp .env.example .env # fill in your keys
ngrok http 3040 # in another terminal
python main.py +1234567890 # Replace with the phone number the code will call
```
## Tests
Running these tests is not required for executing the application.
It is recommended to run tests after making changes to the codebase.
```bash
python -m pytest tests/ -v # runs in ~0.03s
```
## License
MIT
Excerpt of 2,505 characters
Read on GitHub
30
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:bd48b0915a5c3ec1, desc:agent orchestration