Metadata-Version: 2.5
Name: livekit-plugins-nvidia
Version: 1.8.12
Summary: LiveKit Agents Plugin for services from nvidia
Project-URL: Documentation, https://docs.livekit.io
Project-URL: Website, https://livekit.io/
Project-URL: Source, https://github.com/livekit/agents
Author-email: LiveKit <hello@livekit.io>
License-Expression: Apache-2.0
Keywords: audio,livekit,nvidia,realtime,video,webrtc
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.10
Classifier: Topic :: Multimedia :: Sound/Audio
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10.0
Requires-Dist: grpcio>=1.51.1
Requires-Dist: livekit-agents>=1.8.12
Requires-Dist: nvidia-riva-client>=2.16
Provides-Extra: personaplex
Requires-Dist: livekit-agents[codecs]>=1.8.12; extra == 'personaplex'
Requires-Dist: sphn>=0.2.0; extra == 'personaplex'
Description-Content-Type: text/markdown

<!-- SPDX-FileCopyrightText: Copyright (c) 2026, NVIDIA CORPORATION & AFFILIATES. -->
<!-- SPDX-FileCopyrightText: Copyright 2025 LiveKit, Inc. -->
<!-- SPDX-License-Identifier: Apache-2.0 -->

# NVIDIA plugin for LiveKit Agents

Support for NVIDIA Speech AI services in LiveKit Agents.

## Installation

```bash
pip install livekit-plugins-nvidia
```

## Pre-requisites

You can either:

1. Use an API key from NVIDIA. It can be set as an environment variable: `NVIDIA_API_KEY`
2. Use your self-hosted [NIM](https://developer.nvidia.com/nim) server.

## Usage

```python
from livekit.plugins import nvidia

stt = nvidia.STT(
    model="parakeet-1.1b-en-US-asr-streaming-silero-vad-sortformer",
    inference_mode="streaming",
    server="grpc.nvcf.nvidia.com:443",
    use_ssl=True,
    endpointing=nvidia.EndpointingConfig(mode="low_latency"),
)

tts = nvidia.TTS(
    voice="Magpie-Multilingual.EN-US.Leo",
    sample_rate=16000,
    inference_mode="online",
)
```

STT defaults to `inference_mode="auto"`; use `"streaming"` or `"offline"` to
match the deployed model. TTS defaults to `"online"`; use `"offline"` for batch
models.

For local NVIDIA Speech deployments, pass the local `server` and set
`use_ssl=False` when TLS is not enabled.
