Metadata-Version: 2.3
Name: evdm
Version: 0.3.1
Summary: Prototype of a Hierarchical Event Bus for building event driven speech conversational AI systems
Requires-Python: >=3.11
Requires-Dist: aioconsole>=0.8.0
Requires-Dist: deepgram-sdk>=3.7.3
Requires-Dist: loguru>=0.7.2
Requires-Dist: numpy>=2.1.0
Requires-Dist: ollama>=0.3.2
Requires-Dist: pyaudio>=0.2.14
Requires-Dist: sounddevice>=0.5.0
Requires-Dist: soundfile
Requires-Dist: websockets>=13.1
Description-Content-Type: text/markdown

# evdm

![GitHub Tag](https://img.shields.io/github/v/tag/lepisma/evdm)

Prototype of a [Hierarchical Event
Bus](https://lepisma.xyz/2024/08/01/hierarchical-event-bus-for-spoken-conversational-systems/index.html)
for building event driven speech conversational AI systems. Read the linked blog
for more details.

Although it's very simple and direct, you can play around with an example
voicebot built using OpenAI's realtime API in the `./scripts/` directory. After
`uv install`, run `uv run python
./scripts/example-oai-speech-to-speech.py`. Echo cancellation is not available
yet so you should run this example with headphones on. Another example
(`example-oai-dg-multi-speaker.py`) is a multi user voicebot that can handle
talking to multiple speakers.
