Deepgram

The DeepgramTTSService provides ultra-low latency text-to-speech using the Deepgram Aura model. Engineered specifically for conversational AI, it offers human-like speed and quality.

Installation

To use Deepgram TTS, install the required dependencies:

pip install "piopiy-ai[deepgram]"

Prerequisites

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

  export DEEPGRAM_API_KEY="your_api_key_here"

Configuration

DeepgramTTSService Parameters

Parameter	Type	Default	Description
api_key	str	Required	Your Deepgram API key.
voice	str	"aura-2-helena-en"	Deepgram voice/model ID.
encoding	str	"linear16"	Audio encoding format.
sample_rate	int	None	Audio sample rate in Hz.

Usage

Basic Setup (WebSocket)

import os
from piopiy.services.deepgram.tts import DeepgramTTSService

tts = DeepgramTTSService(
    api_key=os.getenv("DEEPGRAM_API_KEY"),
    voice="aura-2-luna-en"
)

Notes

• Conversational Latency: Deepgram Aura is one of the fastest TTS models available, making it excellent for fast-paced bot interactions.
• Flush Command: When using the WebSocket service, Piopiy automatically sends a Flush command when the LLM finishes its response to ensure the entire message is synthesized and sent immediately.
• Interruption: Supports the Clear command via WebSocket to immediately stop audio generation when a user interrupts the bot.