# ElevenLabs Turbo v2.5 - Core.Today AI API

> High-quality, low-latency ElevenLabs text-to-speech in 32 languages. The same 26 premium voices as v3 at half the price, optimized for real-time and high-volume use.

- **Provider**: ElevenLabs
- **Model ID**: elevenlabs/turbo-v2.5
- **Category**: Audio & TTS
- **Credits**: 119 per 1000 characters (0.113 credits/char, billed per character)
- **Speed**: Fast
- **Quality**: High

## Features
- Low latency optimized for real-time apps
- 32 language support
- 26 premium preset voices
- Half the price of ElevenLabs v3
- Stability and similarity controls

## Use Cases
- Real-time voice assistants and chatbots
- High-volume content narration
- IVR and phone system prompts
- Live translation voiceovers
- Interactive game dialogue

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - The text to convert to speech

### Optional
- **speed**: number (default: 1) - Speed of speech (0.7-1.2)
- **style**: number (default: 0) - Style exaggeration (0.0-1.0)
- **voice**: string (default: Rachel) - Voice choice for speech generation (same 26-voice lineup as v3)
  Options: Rachel, Drew, Clyde, Paul, Aria, Domi, Dave, Roger, Fin, Sarah, James, Jane, Juniper, Arabella, Hope, Bradford, Reginald, Gaming, Austin, Kuon, Blondie, Priyanka, Alexandra, Monika, Mark, Grimblewood
- **next_text**: string - Next text for context
- **stability**: number (default: 0.5) - Voice stability (0.0-1.0). Lower is more variable and expressive
- **language_code**: string (default: en) - Language of the text. Match it to your input for correct pronunciation
  Options: en, ko, ja, zh, de, hi, fr, pt, it, es, id, nl, tr, fil, pl, sv, bg, ro, ar, cs, el, fi, hr, ms, sk, da, ta, uk, ru, hu, no, vi
- **previous_text**: string - Previous text for context
- **similarity_boost**: number (default: 0.75) - How closely the output follows the original voice (0.0-1.0)

## Examples

### Chatbot Voice Response
Fast voice response for an interactive assistant
```json
{
  "model": "elevenlabs/turbo-v2.5",
  "input": {
    "prompt": "안녕하세요! 주문하신 상품이 내일 오후에 도착할 예정입니다. 더 도와드릴 일이 있을까요?",
    "voice": "Sarah",
    "language_code": "ko",
    "speed": 1.05
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Best price/performance in the ElevenLabs lineup — ideal default for production TTS
- Set language_code to match your text for correct pronunciation across 32 languages
- Slightly faster speed (1.05-1.1) makes chatbot replies feel more responsive
- For maximum expressiveness with audio tags, use ElevenLabs v3 instead

## Documentation
https://replicate.com/elevenlabs/turbo-v2.5