Gradium STT

The GradiumSTTService provides real-time speech transcription using Gradium's lightning-fast WebSocket API. It is designed for high-performance audio processing with minimal latency.

Installation

To use Gradium, install the required dependencies:

pip install "piopiy-ai[gradium]"

Prerequisites

• A Gradium account and API key (Get yours here).
• Set your API key in your environment:

  export GRADIUM_API_KEY="your_api_key_here"

Configuration

GradiumSTTService Parameters

Parameter	Type	Default	Description
api_key	str	Required	Your Gradium API key.
api_endpoint_base_url	str	"wss://eu.api.gradium.ai/api/speech/asr"	Gradium's streaming endpoint URL.
json_config	str	None	Optional JSON configuration string for additional model settings.

Usage

Basic Setup

import os
from piopiy.services.gradium.stt import GradiumSTTService

stt = GradiumSTTService(
    api_key=os.getenv("GRADIUM_API_KEY")
)

With Custom Endpoint

import os
from piopiy.services.gradium.stt import GradiumSTTService

stt = GradiumSTTService(
    api_key=os.getenv("GRADIUM_API_KEY"),
    api_endpoint_base_url="wss://your-custom-endpoint.gradium.ai/api/speech/asr"
)

Notes

• Sample Rate: The Gradium STT service is optimized for a hardcoded sample rate of 24000 Hz.
• WebSocket Streaming: This service establishes a persistent WebSocket connection to Gradium for real-time transcription.