Nvidia STT
The Nvidia STT provider enables your agent to use Nvidia's Riva speech-to-text models for high-performance, low-latency speech recognition.
Installation
Install the Nvidia-enabled VideoSDK Agents package:
pip install "videosdk-plugins-nvidia"
Authentication
The Nvidia plugin requires an Nvidia API key.
Set NVIDIA_API_KEY in your .env file.
Importing
from videosdk.plugins.nvidia import NvidiaSTT
Example Usage
from videosdk.plugins.nvidia import NvidiaSTT
from videosdk.agents import CascadingPipeline
# Initialize the Nvidia STT model
stt = NvidiaSTT(
# When NVIDIA_API_KEY is set in .env - DON'T pass api_key parameter
api_key="your-nvidia-api-key",
model="parakeet-1.1b-en-US-asr-streaming-silero-vad-sortformer",
language_code="en-US",
profanity_filter=False,
automatic_punctuation=True
)
# Add stt to cascading pipeline
pipeline = CascadingPipeline(stt=stt)
note
When using .env file for credentials, don't pass them as arguments to model instances or context objects. The SDK automatically reads environment variables, so omit api_key and other credential parameters from your code.
Configuration Options
api_key: Your Nvidia API key (required, can also be set via environment variable)model: The Nvidia Riva model to use (default:"parakeet-1.1b-en-US-asr-streaming-silero-vad-sortformer")server: The Nvidia Riva server address (default:"grpc.nvcf.nvidia.com:443")function_id: The specific function ID for the service (default:"1598d209-5e27-4d3c-8079-4751568b1081")language_code: Language code for transcription (default:"en-US")sample_rate: Audio sample rate in Hz (default:16000)profanity_filter: (bool) Enable or disable profanity filtering (default:False)automatic_punctuation: (bool) Enable or disable automatic punctuation (default:True)use_ssl: (bool) Enable SSL connection (default:True)
Additional Resources
The following resources provide more information about using Nvidia Riva with VideoSDK Agents SDK.
- Nvidia Riva docs: Nvidia Riva documentation.
Got a Question? Ask us on discord

