LiveKit w/ AssemblyAI, Cerebras, and Rime Example Project

A sample LiveKit Agent using AssemblyAI for handling STT (Speech-To-Text), Cerebras as the agent model, and Rime for handling TTS (Text-To-Speech).

Prerequisites

• Python 3.13+
• Linux, MacOS, or Windows
• LiveKit Cloud Account
• AssemblyAI Account
• Cerebras Account
• Rime Account

Get Started

Clone this repo then:

1. Set up your Python environment

A virtual environment is recommended to manage your Python dependencies. Using (uv)[https://docs.astral.sh/uv/]:

# Create virtual environment and download project dependencies
uv sync

2. Create a .env file

Copy the env.sample file to a .env file with:

cp env.example .env

Then fill in the API key values.

3. Run the agent

Locally:

uv run python agent.py console

Via LiveKit Playground:

uv run python agent.py dev

Documentation & Resources

For more details on LiveKit:

• LiveKit Agents Docs
• LiveKit Slack Channel
• LiveKit Repos

Cerebras:

• Cerebras Docs
• LiveKit + Cerebras Docs

Rime:

• Rime Docs
• LiveKit + Rime Docs

For AssemblyAI:

• LiveKit + AssemblyAI Docs
• AssemblyAI Universal Streaming
• AssemblyAI's Developer Documentation
• AssemblyAI Discord Community
• AssemblyAI Youtube Channel