Enterprise Plans
AI Application Development
AI Translator API
AI SEO/GEO Service
Web Scraping Service
Log in
const fs = require('fs');
const path = require('path');
const axios = require('axios').default;
const api = new axios.create({
baseURL: 'https://api.ai.cc/v1',
headers: { Authorization: 'Bearer ' },
});
const main = async () => {
const response = await api.post(
'/tts',
{
model: '#g1_aura-asteria-en',
text: 'Hi! What are you doing today?',
},
{ responseType: 'stream' },
);
const dist = path.resolve(__dirname, './audio.wav');
const writeStream = fs.createWriteStream(dist);
response.data.pipe(writeStream);
writeStream.on('close', () => console.log('Audio saved to:', dist));
};
main();
import os
import requests
def main():
url = "https://api.ai.cc/v1/tts"
headers = {
"Authorization": "Bearer ",
}
payload = {
"model": "#g1_aura-asteria-en",
"text": "Hi! What are you doing today?",
}
response = requests.post(url, headers=headers, json=payload, stream=True)
dist = os.path.join(os.path.dirname(__file__), "audio.wav")
with open(dist, "wb") as write_stream:
for chunk in response.iter_content(chunk_size=8192):
if chunk:
write_stream.write(chunk)
print("Audio saved to:", dist)
main()
-
AI Playground
Test all API models in the sandbox environment before you integrate.
We provide more than 300 models to integrate into your app.
.png)
Product Detail
Deepgram Aura: Revolutionizing Real-time Text-to-Speech
Experience the next generation of conversational AI with Deepgram Aura. Launched in June 2023, Aura is a groundbreaking text-to-speech (TTS) AI model engineered for the demands of real-time, interactive applications. It sets a new standard for human-like voice quality, unparalleled speed, and remarkable efficiency, transforming how developers build responsive voice AI experiences.
Developer: Deepgram
Release Date: June 2023
Model Type: Text-to-speech (TTS)
Version: 1.0 (Initial Release)
✨Key Features & Advantages
- ✔️ Natural, Low-Latency Voices: Access a diverse range of natural, human-like voices with significantly lower latency compared to any alternative voice AI.
- 🚀 Optimized for Conversational AI: Specifically designed for responsive, real-time conversational AI agents and applications, ensuring smooth and dynamic interactions.
- 🔗 Seamless Integration: Effortlessly integrates with Deepgram's industry-leading Nova Speech-to-Text API, creating a complete voice AI solution.
- 🌐 Extensive Language Support: Capable of handling multiple languages, accents, and dialects with exceptional accuracy and fluidity.
💡Intended Use Cases
Deepgram Aura excels in scenarios demanding high-throughput voice interactions and responsive AI agents. It is the ideal choice for:
- • Customer Service: Powering advanced virtual assistants for improved customer experience.
- • Virtual Assistants: Creating highly engaging and lifelike digital assistants.
- • Interactive Voice Response (IVR) Systems: Delivering natural and efficient automated phone interactions.
🗣️Available Aura Voices
Deepgram Aura offers a variety of distinct English voices to suit different applications and preferences:
- Asteria: `#g1_aura-asteria-en`
- Hera: `#g1_aura-hera-en`
- Luna: `#g1_aura-luna-en`
- Stella: `#g1_aura-stella-en`
- Athena: `#g1_aura-athena-en`
- Zeus: `#g1_aura-zeus-en`
- Orion: `#g1_aura-orion-en`
- Arcas: `#g1_aura-arcas-en`
- Perseus: `#g1_aura-perseus-en`
- Angus: `#g1_aura-angus-en`
- Orpheus: `#g1_aura-orpheus-en`
- Helios: `#g1_aura-helios-en`
⚙️Technical Deep Dive into Aura
Architecture & Performance
Aura's architecture is a testament to Deepgram's leadership in speech technology. Optimized for unmatched speed and efficiency, it stands as the fastest high-quality TTS option available. This innovation stems from Deepgram's extensive expertise in processing and modeling speech audio, particularly for real-time streaming use cases when paired with their STT models.
Performance Metrics: Aura delivers unrivaled performance, boasting lower latency and superior voice quality compared to any other comparable TTS model. Its real-world effectiveness is validated by successful deployments with numerous Deepgram customers.
Training Data & Data Source
Aura has been rigorously trained on millions of hours of high-quality audio data. This extensive training enables it to produce natural-sounding voices across diverse languages and domains. Deepgram has meticulously curated a diverse dataset, ensuring Aura's robust performance across a wide array of use cases. The model's knowledge cutoff aligns with its initial release in June 2023.
Diversity & Bias Mitigation: Deepgram prioritizes responsible AI development. Significant steps have been taken to train Aura on diverse data, actively minimizing potential biases and ensuring equitable performance across various demographics and use cases.
📊Aura vs. Other TTS Models
Deepgram Aura consistently outperforms other popular TTS models across several critical dimensions:
- ★ Superior Voice Quality: Delivers exceptionally natural-sounding, human-like voices with high accuracy and minimal errors.
- ★ Unrivaled Responsiveness: As the fastest high-quality TTS option, it boasts low latency and high throughput, crucial for real-time interactions.
- ★ Cost-Efficiency & Adaptability: Designed to handle diverse inputs and seamlessly adapt to various accents, dialects, and use cases, providing excellent value.
🚀Getting Started with Deepgram Aura
API Usage Example
Integrating Deepgram Aura into your applications is straightforward. Here’s a typical API usage snippet (placeholder for actual code example):
<snippet data-name="voice.tts" data-model="#g1_aura-asteria-en"></snippet>
For detailed documentation and more examples, please refer to the official Deepgram API documentation.
Ethical Guidelines & Licensing
Deepgram is deeply committed to responsible AI development. Ethical considerations are integrated into Aura's design and deployment lifecycle. The company actively expands Aura's capabilities, with continuous plans for more lifelike voices, additional language support, and innovative new features to ensure a positive impact.
License Type: Deepgram Aura is available under a flexible licensing model, supporting both commercial and non-commercial use, with pricing scaled based on usage.
❓Frequently Asked Questions (FAQs)
A: Deepgram Aura is a state-of-the-art Text-to-Speech (TTS) AI model designed for real-time, conversational AI applications, offering human-like voice quality at high speed and efficiency.
A: Aura stands out with its exceptionally low latency, superior voice quality, and optimization specifically for responsive conversational AI, making it ideal for interactive real-time experiences.
A: Aura is best suited for applications requiring high-throughput voice interactions, such as customer service virtual assistants, general virtual assistants, and advanced Interactive Voice Response (IVR) systems.
A: Yes, Aura supports multiple languages and is designed to handle diverse accents and dialects effectively, ensuring broad global applicability.
A: Deepgram Aura is accessible via API, with detailed documentation available. It offers flexible licensing for both commercial and non-commercial use, priced based on usage.
Learn how you can transformyour company with AICC APIs
