livekit-plugins-rime

What it is and what it does

This package is a plugin that adds Rime speech-to-text (STT) capability to LiveKit Agents, the framework for building real-time voice agents that run on servers. LiveKit Agents lets you compose conversational AI by mixing STT, LLM, and TTS providers; this plugin supplies Rime as the speech recognition layer. You install it alongside livekit-agents and use it to instantiate a Rime STT provider in your agent session, allowing your voice agent to understand spoken input from users in real-time conversations.

The plugin is part of LiveKit's broader ecosystem for building multi-modal agents that can see, hear, and respond. It integrates with the framework's job scheduling, WebRTC clients, and agent handoff patterns. Since it released recently (January 2025) and is actively maintained, it represents a current integration point for Rime speech recognition within the LiveKit platform.

Use it for:

• Build a voice agent that uses Rime for speech recognition instead of Deepgram or other STT providers.
• Replace or supplement an existing STT provider in a multi-turn conversational agent without rewriting core logic.
• Deploy a real-time voice assistant on LiveKit infrastructure with Rime as the hearing layer.
• Test or evaluate Rime's speech recognition quality within a production-ready agent framework.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Integrates Rime speech recognition into LiveKit's agent framework for building real-time voice agents that can hear and respond to users.

Yes—if you are already building with LiveKit Agents and want to use Rime for speech recognition. The plugin has low install friction, permissive licensing, no known vulnerabilities, and sits on an actively maintained framework with strong community adoption. Install it only if Rime is your preferred STT provider or you are evaluating it; if you are starting fresh, compare Rime's capabilities and cost against other STT options in the LiveKit ecosystem first.

Install

livekit-plugins-rime on PyPI

pip install livekit-plugins-rime

uv add livekit-plugins-rime

poetry add livekit-plugins-rime

Installing livekit-plugins-rime

Before you install

Low friction install with a single runtime dependency on livekit-agents. Package released 1 day ago and actively maintained with 13003 repository stars, indicating a mature underlying framework.

License in practice

Licensed under Apache-2.0 (permissive), allowing commercial and private use with minimal restrictions—suitable for most production deployments.

Quickstart

pip install livekit-plugins-rime

from livekit.agents import Agent, AgentSession
from livekit.plugins import rime

session = AgentSession(
    stt=rime.STT(),  # Rime speech-to-text provider
)
await session.start(agent=agent, room=ctx.room)

Requires livekit-agents as a runtime dependency and a LiveKit server connection (LIVEKIT_URL, LIVEKIT_API_KEY, LIVEKIT_API_SECRET environment variables).

Verify before relying

• Specific Rime model variants or configuration options supported by this plugin version.
• Whether Rime STT supports language variants beyond what the base framework documents.
• Performance characteristics (latency, accuracy) compared to other STT providers in the LiveKit ecosystem.

Package facts

Evidence: livekit_plugins_rime-1.6.10-py3-none-any.whl

Keywords: audio, livekit, realtime, rime, video, webrtc

Tags

More Artificial Intelligence packages

Further reading