# Chatterbox - Core.Today AI API

> Resemble AI's production-grade open-source TTS with unique emotion exaggeration control and instant voice cloning from a short reference audio. MIT-licensed and benchmarked against leading closed-source systems.

- **Provider**: Resemble AI
- **Model ID**: resemble-ai/chatterbox
- **Category**: Audio & TTS
- **Credits**: 59 per 1000 characters (0.057 credits/char, billed per character)
- **Speed**: Fast
- **Quality**: High

## Features
- Unique emotion exaggeration control
- Instant voice cloning from short audio
- Production-grade open-source quality
- CFG/pace control for delivery speed
- Very low per-character price

## Use Cases
- Emotive character voices
- Voice cloning experiments
- Cost-efficient narration
- Dialogue prototyping for games
- Personalized audio content

## 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 - Text to synthesize

### Optional
- **audio_prompt**: string - Reference audio file for voice cloning (optional)
- **exaggeration**: number (default: 0.5) - Emotion exaggeration (0.25-2.0). Neutral = 0.5; extreme values may be unstable
- **cfg_weight**: number (default: 0.5) - CFG/pace weight (0.2-1.0)
- **temperature**: number (default: 0.8) - Controls randomness (0.05-5.0)
- **seed**: integer (default: 0) - Seed (0 for random)

## Examples

### Emotive Announcement
Expressive delivery with raised exaggeration
```json
{
  "model": "resemble-ai/chatterbox",
  "input": {
    "prompt": "We're excited to introduce Chatterbox, our first production-grade open source TTS model!",
    "exaggeration": 0.8,
    "cfg_weight": 0.5
  }
}
```

## 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
- exaggeration 0.5 is neutral — raise toward 1.0 for dramatic delivery, keep under 1.5 for stability
- Lower cfg_weight slows and relaxes the pace; raise it for tighter delivery
- Provide clean reference audio in audio_prompt to clone a voice instantly
- For 23-language support use Chatterbox Multilingual; for speed use Chatterbox Turbo

## Documentation
https://replicate.com/resemble-ai/chatterbox