thinnestAI's text-to-speech engine — 5 models, 90+ voices, 20+ languages with ultra-low latency.

Aero TTS

Aero TTS is thinnestAI's built-in text-to-speech engine. Convert any text into natural, human-sounding speech with ultra-low latency — perfect for voice agents, IVR systems, accessibility features, and content creation.

Why Aero TTS?

Ultra-low latency — ~60-120ms time-to-first-byte depending on model
90+ voices — Male and female voices across 20+ languages
3 model tiers — From premium high-quality to lightweight cost-effective models
Voice mixing — Blend multiple voices to create unique custom voices
Streaming — Start playback before synthesis completes
Simple API — One endpoint, one API key, any programming language
Pay-as-you-go — Billed per character with generous free tier

Models

Aero TTS offers 3 models optimized for different use cases:

Model	Latency	Quality	Best For
aero	~100ms	Highest	Production voice agents, phone calls
aero-mini	~200ms	Medium	Cost-effective batch processing
aero-nano	~150ms	Medium	Lightweight, low-resource environments

Choosing a Model

Voice agents / phone calls — Use aero for the most natural, lowest-latency output
High-volume batch jobs — Use aero-mini for cost-effective processing
Lightweight usage — Use aero-nano for minimal resource usage

Supported Languages

Language	Code	Models
English (US)	`en-us`	aero, aero-mini, aero-nano
English (UK)	`en-gb`	aero, aero-mini, aero-nano
Hindi	`hi` / `hi-in`	aero
Spanish	`es`	aero-mini, aero-nano
French	`fr`	aero-mini, aero-nano
German	`de`	aero-mini, aero-nano
Italian	`it`	aero-mini, aero-nano
Chinese	`zh`	aero-mini, aero-nano
Japanese	`ja`	aero-mini, aero-nano
Korean	`ko`	aero-mini, aero-nano
Portuguese	`pt`	aero-mini, aero-nano

Quick Start

1. Get an API Key

Go to the Aero TTS page in your thinnestAI dashboard and click API Keys > Create Key.

2. Synthesize Speech

curl -X POST https://api.thinnest.ai/api/tts/synthesize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello! Welcome to Aero TTS.",
    "voice": "aero-vayu",
    "model": "aero",
    "format": "mp3"
  }' \
  --output speech.mp3

3. Play the Audio

# macOS
afplay speech.mp3

# Linux
aplay speech.mp3

# Windows
start speech.mp3

That's it. You now have natural-sounding speech from text.

Audio Formats

Format	Extension	Use Case
`mp3`	`.mp3`	Web playback, general use (default)
`wav`	`.wav`	Lossless, professional audio
`pcm`	`.pcm`	Raw audio for real-time pipelines
`opus`	`.opus`	Low-bandwidth streaming
`ogg`	`.ogg`	Open-source applications
`flac`	`.flac`	Lossless compression
`aac`	`.aac`	Mobile and Apple ecosystem

Dashboard

The Aero TTS dashboard in thinnestAI provides:

Playground — Type text and hear it spoken instantly. Compare voices and models side by side.
Voices — Browse all 90+ voices, filter by language and gender, preview samples.
Voice Lab — Create custom hybrid voices by blending 2-4 voices together.
API Keys — Create, manage, and revoke API keys with per-key rate limits.
Usage — Track character usage, request counts, latency metrics, and costs.

Next Steps

Voices — Browse all available voices by language and gender
API Reference — Full REST API documentation with examples
Voice Lab — Create custom hybrid voices
Pricing — Understand pricing tiers and free allowances
SDKs & Integration — Python, JavaScript, and cURL examples

Aero TTS

On this page