# Clova Voice TTS Premium - Core.Today AI API

> NAVER Clova Voice Premium TTS with 108 voices across 6 languages. High-quality Korean voice synthesis with emotion control, Pro voices, and bilingual support.

- **Provider**: NCP Clova
- **Model ID**: ncp-clova/tts-premium
- **Category**: Audio & TTS
- **Credits**: 153 per 1,000 characters (0.15 credits/char, billed by exact character count)
- **Speed**: Fast
- **Quality**: High

## Features
- 108 voice options across 6 languages
- High-quality Korean voice synthesis
- Emotion control (neutral, sad, happy, angry)
- 9 Pro-quality premium voices
- Bilingual Korean+English voices
- Volume, speed, pitch fine-tuning

## Use Cases
- Korean content narration
- Multilingual customer service
- E-learning and educational content
- Interactive voice response (IVR)
- Audiobook production
- Character voice acting

## 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
- **text**: string - Text to synthesize (UTF-8). Korean/Chinese/Japanese/Taiwanese max 2,000 chars, English/Spanish max 3,000 chars.
- **speaker**: string - Voice ID. 108 voices: 81 Korean, 15 Japanese, 4 English, 2 Korean+English, 2 Chinese, 2 Spanish, 2 Taiwanese.
  Options: dara_ang, jinho, mijin, napple, nara_call, nara, nbora, ndaeseong, ndain, ndonghyun, nes_c_hyeri, nes_c_kihyo, nes_c_mikyung, nes_c_sohyun, neunseo, neunwoo, neunyoung, ngaram, ngoeun, ngyeongjun, nhajun, nheera, nian, nihyun, njaewook, njangj, njihun, njihwan, njinho, njiwon, njiyun, njonghyeok, njonghyun, njooahn, njoonyoung, nkitae, nkyunglee, nkyungtae, nkyuwon, nmammon, nmeow, nmijin, nminjeong, nminsang, nminseo, nminyoung, nmovie, noyj, nraewon, nreview, nsabina, nsangdo, nseonghoon, nseungpyo, nshasha, nsinu, nsiyoon, nsujin, nsunhee, nsunkyung, ntaejin, ntiffany, nwontak, nwoof, nwoosik, nyeji, nyejin, nyounghwa, nyoungil, nyoungmi, nyujin, nyuna, vara, vdaeseong, vdain, vdonghyun, vgoeun, vhyeri, vian, vmikyung, vyuna, dara-danna, dsinu-matt, liangliang, meimei, dayumu, ddaiki, deriko, dhajime, dmio, dnaomi, dnaomi_formal, dnaomi_joyful, driko, dsayuri, dtomoko, nnaomi, nsayuri, ntomoko, shinji, clara, danna, djoey, matt, carmen, jose, chiahua, kuanlin

### Optional
- **volume**: integer (default: 0) - Volume (-5 to 5, default 0)
- **speed**: integer (default: 0) - Speed (-5 to 10, default 0)
- **pitch**: integer (default: 0) - Pitch (-5 to 5, default 0)
- **emotion**: integer (default: 0) - Emotion (0: neutral, 1: sad, 2: happy, 3: angry). Supported by select voices only.
  Options: 0, 1, 2, 3
- **emotion-strength**: integer (default: 1) - Emotion strength (0: weak, 1: normal, 2: strong)
  Options: 0, 1, 2
- **format**: string (default: mp3) - Audio output format
  Options: mp3, wav

## Examples

### Korean TTS with Emotion
Generate Korean speech with happy emotion using the default 'nara' voice
```json
{
  "model": "ncp-clova/tts-premium",
  "input": {
    "text": "안녕하세요, 코어닷투데이입니다. 최고의 AI 서비스를 경험해 보세요!",
    "speaker": "vara",
    "emotion": 2,
    "emotion-strength": 2,
    "speed": 0,
    "format": "mp3"
  }
}
```

## 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
- Use Pro voices (vara, vgoeun, etc.) for the highest quality Korean speech
- Emotion control is only supported by select Korean voices - test with nara first
- For Korean+English bilingual content, use dara-danna (female) or dsinu-matt (male)
- Keep text under 2,000 characters for Korean/Chinese/Japanese and 3,000 for English/Spanish

## Documentation
https://api.ncloud-docs.com/docs/ai-naver-clovavoice-ttspremium