# Gemini 3.8 Flash TTS

> Generate expressive multi-speaker speech with Gemini 3.8 Flash TTS. Studio-grade voice fidelity and long-form stability. Try free, pay per use.

## 개요

- **Endpoint**: `POST https://api.muapi.ai/api/v1/gemini-3-8-flash-tts`
- **모델 ID**: `gemini-3-8-flash-tts`
- **카테고리**: AI 오디오
- **변형**: 3.8 Flash
- **패밀리**: gemini-tts
- **비용**: 호출당 0.015 크레딧(일부 모델은 매개변수에 따라 동적으로 비용을 계산함)

## API 사용법

Muapi는 **제출 후 폴링(Submit-then-poll)** 패턴을 사용합니다. 즉, 작업을 제출하고 `request_id`를 얻은 다음 `status`가 `completed`가 될 때까지 엔드포인트를 폴링합니다. 선택적으로 제출 호출 시 `?webhook=YOUR_URL`을 전달하여 작업 완료 시 POST 웹훅 콜백을 수신할 수 있습니다(폴링 생략 가능).

**인증**: `x-api-key` 헤더에 MuApi 키를 보냅니다. https://muapi.ai/access-keys에서 하나 받으세요.

### 1. 작업 제출

```http
POST https://api.muapi.ai/api/v1/gemini-3-8-flash-tts
Content-Type: application/json
x-api-key: YOUR_API_KEY
```

**최소(필수):**

```json
{
  "speakers": [
    {
      "speaker_id": "Speaker 1",
      "voice_name": "Fenrir",
      "audio_profile": "A stern and weary gatekeeper",
      "accent": "British (RP)",
      "style": "Deadpan",
      "pace": "Natural"
    },
    {
      "speaker_id": "Speaker 2",
      "voice_name": "Puck",
      "audio_profile": "A determined and courageous traveler seeking answers.",
      "accent": "American (Gen)",
      "style": "Empathetic",
      "pace": "Staccato"
    }
  ],
  "dialogue_turns": [
    {
      "speaker_id": "Speaker 1",
      "text": "[shouting] Halt, traveler! The northern pass is sealed by order of the council."
    },
    {
      "speaker_id": "Speaker 2",
      "text": "[determination] I carry a message for the elder. Step aside, or I will force my way through."
    },
    {
      "speaker_id": "Speaker 1",
      "text": "[caution] No one passes. [pensive] The elder is... he's no longer receiving visitors."
    },
    {
      "speaker_id": "Speaker 2",
      "text": "It's too late. [whispers] The shadow... it reached him first. [urgency] You need to leave. [shouting] Now."
    }
  ]
}
```

**전체 예(모든 매개변수):**

```json
{
  "speakers": [
    {
      "speaker_id": "Speaker 1",
      "voice_name": "Fenrir",
      "audio_profile": "A stern and weary gatekeeper",
      "accent": "British (RP)",
      "style": "Deadpan",
      "pace": "Natural"
    },
    {
      "speaker_id": "Speaker 2",
      "voice_name": "Puck",
      "audio_profile": "A determined and courageous traveler seeking answers.",
      "accent": "American (Gen)",
      "style": "Empathetic",
      "pace": "Staccato"
    }
  ],
  "dialogue_turns": [
    {
      "speaker_id": "Speaker 1",
      "text": "[shouting] Halt, traveler! The northern pass is sealed by order of the council."
    },
    {
      "speaker_id": "Speaker 2",
      "text": "[determination] I carry a message for the elder. Step aside, or I will force my way through."
    },
    {
      "speaker_id": "Speaker 1",
      "text": "[caution] No one passes. [pensive] The elder is... he's no longer receiving visitors."
    },
    {
      "speaker_id": "Speaker 2",
      "text": "It's too late. [whispers] The shadow... it reached him first. [urgency] You need to leave. [shouting] Now."
    }
  ],
  "scene": "",
  "sample_context": "",
  "temperature": 1
}
```

**응답:**

```json
{
  "request_id": "abc123",
  "status": "processing"
}
```

### 2. 결과 폴링

```http
GET https://api.muapi.ai/api/v1/predictions/{request_id}/result
x-api-key: YOUR_API_KEY
```

가능한 'status' 값: 'queued', 'pending', 'processing', 'completed', 'failed', 'cancelled'. 2~5초마다 폴링을 수행하세요. `completed` 상태가 되면 결과 URL이 `outputs` 배열에 반환됩니다.

**`completed` 시 응답 예시:**

```json
{
  "id": "abc123",
  "status": "completed",
  "outputs": [
    "https://cdn.muapi.ai/.../output.png"
  ],
  "urls": {
    "get": "https://api.muapi.ai/api/v1/predictions/abc123/result"
  },
  "created_at": "2026-05-08T12:34:56Z",
  "has_nsfw_contents": []
}
```

### cURL

```bash
1. 작업 제출
REQUEST_ID=$(curl -s -X POST https://api.muapi.ai/api/v1/gemini-3-8-flash-tts \
  -H "x-api-key: $MUAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"speakers":[{"speaker_id":"Speaker 1","voice_name":"Fenrir","audio_profile":"A stern and weary gatekeeper","accent":"British (RP)","style":"Deadpan","pace":"Natural"},{"speaker_id":"Speaker 2","voice_name":"Puck","audio_profile":"A determined and courageous traveler seeking answers.","accent":"American (Gen)","style":"Empathetic","pace":"Staccato"}],"dialogue_turns":[{"speaker_id":"Speaker 1","text":"[shouting] Halt, traveler! The northern pass is sealed by order of the council."},{"speaker_id":"Speaker 2","text":"[determination] I carry a message for the elder. Step aside, or I will force my way through."},{"speaker_id":"Speaker 1","text":"[caution] No one passes. [pensive] The elder is... he's no longer receiving visitors."},{"speaker_id":"Speaker 2","text":"It's too late. [whispers] The shadow... it reached him first. [urgency] You need to leave. [shouting] Now."}]}' | jq -r .request_id)

2. 완료될 때까지 폴링
while :; do
  RESP=$(curl -s https://api.muapi.ai/api/v1/predictions/$REQUEST_ID/result -H "x-api-key: $MUAPI_API_KEY")
  STATUS=$(echo "$RESP" | jq -r .status)
  [ "$STATUS" = "completed" ] && echo "$RESP" | jq .outputs && break
  [ "$STATUS" = "failed" ] && echo "$RESP" && exit 1
  sleep 3
done
```

### Python

```python
import os, time, requests

API = "https://api.muapi.ai/api/v1"
headers = {"x-api-key": os.environ["MUAPI_API_KEY"]}

r = requests.post(f"{API}/gemini-3-8-flash-tts", headers=headers, json={"speakers":[{"speaker_id":"Speaker 1","voice_name":"Fenrir","audio_profile":"A stern and weary gatekeeper","accent":"British (RP)","style":"Deadpan","pace":"Natural"},{"speaker_id":"Speaker 2","voice_name":"Puck","audio_profile":"A determined and courageous traveler seeking answers.","accent":"American (Gen)","style":"Empathetic","pace":"Staccato"}],"dialogue_turns":[{"speaker_id":"Speaker 1","text":"[shouting] Halt, traveler! The northern pass is sealed by order of the council."},{"speaker_id":"Speaker 2","text":"[determination] I carry a message for the elder. Step aside, or I will force my way through."},{"speaker_id":"Speaker 1","text":"[caution] No one passes. [pensive] The elder is... he's no longer receiving visitors."},{"speaker_id":"Speaker 2","text":"It's too late. [whispers] The shadow... it reached him first. [urgency] You need to leave. [shouting] Now."}]})
request_id = r.json()["request_id"]

while True:
    res = requests.get(f"{API}/predictions/{request_id}/result", headers=headers).json()
    if res["status"] == "completed":
        print(res["outputs"]); break
    if res["status"] == "failed":
        raise RuntimeError(res.get("error")); 
    time.sleep(3)
```

## 입력 스키마

API는 다음 입력 매개변수를 허용합니다.

- **`speakers`** (`array`, _필수_):
  . ID .
  - **`speakers[].speaker_id`** (`string`, _필수_): . "스피커 N"("스피커 1").
  - **`speakers[].voice_name`** (`string`, _필수_): 쌍둥이자리.
    - 옵션: `"Achernar"`, `"Achird"`, `"Algenib"`, `"Algieba"`, `"Alnilam"`, `"Aoede"`, `"Autonoe"`, `"Callirrhoe"`, `"Charon"`, `"Despina"`, `"Enceladus"`, `"Erinome"`, `"Fenrir"`, `"Gacrux"`, `"Iapetus"`, `"Kore"`, `"Laomedeia"`, `"Leda"`, `"Orus"`, `"Puck"`, `"Pulcherrima"`, `"Rasalgethi"`, `"Sadachbia"`, `"Sadaltager"`, `"Schedar"`, `"Sulafat"`, `"Umbriel"`, `"Vindemiatrix"`, `"Zephyr"`, `"Zubenelgenubi"`
  - **`speakers[].audio_profile`** (`string`, _선택사항_): Optional natural-language description of the persona, e.g. "A warm and soothing narrator".
  - **`speakers[].accent`** (`string`, _필수_): .
    - 기본값: `"Neutral"`
    - 옵션: `"Neutral"`, `"American (Gen)"`, `"American (Valley)"`, `"American (South)"`, `"British (RP)"`, `"British (Brixton)"`, `"Transatlantic"`, `"Australian"`
  - **`speakers[].style`** (`string`, _필수_): .
    - 기본값: `"Empathetic"`
    - 옵션: `"Vocal Smile"`, `"Newscaster"`, `"Whisper"`, `"Empathetic"`, `"Promo/Hype"`, `"Deadpan"`
  - **`speakers[].pace`** (`string`, _필수_): .
    - 기본값: `"Natural"`
    - 옵션: `"Natural"`, `"Rapid Fire"`, `"The Drift"`, `"Staccato"`

- **`dialogue_turns`** (`array`, _필수_):
  . 스피커_ID . [소리친다] [속삭인다] .
  - **`dialogue_turns[].speaker_id`** (`string`, _필수_): ID ( "발표자 1" ) .
  - **`dialogue_turns[].text`** (`string`, _필수_): . 와 호환됩니다. 10000 .

- **`scene`** (`string`, _선택사항_):
  Optional scene description that sets the acoustic setting, e.g. "A quiet, warm room with a fireplace crackling softly."
  - 기본값: `""`

- **`sample_context`** (`string`, _선택사항_):
  Optional overall tone/style, e.g. "Audiobook style narration. Tone is gentle and inviting."
  - 기본값: `""`

- **`temperature`** (`number`, _선택사항_):
  (0-2) . , .
  - 기본값: `1`
  - 범위: `0` to `2`

## 출력 스키마

폴링 엔드포인트는 다음 필드를 반환합니다.

- **`id`** (`string`): 요청 ID입니다.
- **`status`** (`string`): 'queued', 'pending', 'processing', 'completed', 'failed', 'cancelled' 중 하나입니다.
- **`outputs`** (`array`): 생성된 이미지/비디오/오디오에 대한 URL 목록입니다. `status`가 `completed`가 될 때까지 비어 있습니다.
- **`urls.get`** (`string`): 이 예측 결과를 다시 조회하는 자체 링크입니다.
- **`error`** (`string` | `null`): `status`가 `failed`인 경우의 오류 메시지입니다.
- **`created_at`** (`string`): 요청이 생성된 시점의 ISO-8601 타임스탬프입니다.
- **`has_nsfw_contents`** (`array of boolean`): 출력별 NSFW 감지 플래그입니다.

## 웹훅(선택사항)

제출 URL에 `?webhook=https://your-server/path`를 추가하세요. 작업이 최종 상태에 도달하면 Muapi가 해당 URL로 폴링 응답과 동일한 페이로드를 POST 전송하므로 별도의 폴링이 필요하지 않습니다.

## 에이전트 연동

Muapi는 MCP 서버와 CLI를 제공하므로 AI 에이전트(Claude Code, Cursor, 커스텀 에이전트)가 복잡한 HTTP 코드 없이도 이 엔드포인트를 호출할 수 있습니다.

```bash
# CLI 설치
npm install -g muapi-cli

# 1회 인증
muapi auth login

# 모든 Muapi 모델을 에이전트에 MCP 도구로 제공합니다.
muapi mcp serve
```

MCP 서버는 `gemini-3-8-flash-tts`을 포함하여 모든 모델에 대해 제출 + 폴링을 래핑하는 도구를 노출합니다. 카테고리별 단축키(`muapi image generate`, `muapi video from-image` 등)는 `muapi --help`를 참조하세요.

## 관련 모델
- [3.1 Flash](https://muapi.ai/ko/playground/gemini-3-1-flash-tts)
- [2.5 Pro](https://muapi.ai/ko/playground/gemini-2-5-pro-tts)
- [3.8 Flash Lite](https://muapi.ai/ko/playground/gemini-3-8-flash-lite-tts)
- [3.8 Flash](https://muapi.ai/ko/playground/gemini-3-8-flash-tts)

## 자원
- [플레이그라운드 페이지](https://muapi.ai/ko/playground/gemini-3-8-flash-tts)
- [API 참조](https://muapi.ai/ko/playground/gemini-3-8-flash-tts/api)
- [전역 llms.txt](https://muapi.ai/ko/llms.txt)