Request Parameters
Complete reference of all parameters for the TTS generate endpoint.
Required Parameters
| Parameter | Type | Description |
|---|
text |
string |
The text to convert to speech. Maximum 5,000 characters. |
Optional Parameters
| Parameter | Type | Default | Description |
|---|
voice |
string |
"Peter" |
Voice name. Use /voices endpoint to get available voices. |
language |
string |
"en-US" |
Language code (e.g., "en-US", "hi", "ja-JP"). |
model |
string |
"aura-prime" |
TTS model: aura-lite, aura-prime, aura-max, rapid-flash, rapid-max |
speed |
float |
1.0 |
Playback speed. Range: 0.5 to 2.0 |
pitch |
float |
1.0 |
Voice pitch. Range: 0.5 to 2.0 |
format |
string |
"mp3" |
Audio format: mp3, wav |
Parameter Details
text
- Maximum 5,000 characters per request
- Supports expression tags:
<laugh>, <sigh>, etc. (Aura models only)
voice
Voice names are localized by language. The same underlying voice has different names:
- English: Peter, Emma, James, Olivia...
- Hindi: Rahul, Deepika, Vikram, Priya...
- Japanese: Takeshi, Yuki, Kenji, Sakura...
model
| Model | Speed | Quality | Expressions |
|---|
aura-lite | Fast | Good | Yes |
aura-prime | Medium | Excellent | Yes |
aura-max | Slow | Premium | Yes |
rapid-flash | Fastest | Good | No |
rapid-max | Fast | Very Good | No |
format
| Format | MIME Type | Notes |
|---|
mp3 | audio/mpeg | Default, good compression |
wav | audio/wav | Uncompressed, best for editing |
Example Requests
Minimal Request
{
"text": "Hello, world!"
}
Full Request
{
"text": "Premium quality audio for professional content.",
"voice": "Peter",
"language": "en-US",
"model": "aura-max",
"speed": 0.9,
"pitch": 1.0,
"format": "wav"
}
With Expressions
{
"text": "That's hilarious! <laugh> I can't believe it!",
"voice": "Peter",
"language": "en-US",
"model": "aura-prime"
}