102 lines
2.7 KiB
Python
102 lines
2.7 KiB
Python
"""한국어 우선 다중 엔진 TTS 웹 서비스.
|
|
|
|
- 로그인/과금 없이 무제한 사용
|
|
- 엔진(MeloTTS, Coqui GlowTTS-KSS 등) 선택 → 언어/목소리 선택
|
|
- 속도 / 피치 조절
|
|
- GPU 자동 사용
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import os
|
|
|
|
from fastapi import FastAPI, HTTPException
|
|
from fastapi.middleware.cors import CORSMiddleware
|
|
from fastapi.responses import FileResponse, Response
|
|
from fastapi.staticfiles import StaticFiles
|
|
from pydantic import BaseModel, Field
|
|
|
|
from .engines import (
|
|
default_engine_id,
|
|
get_engine,
|
|
list_engines,
|
|
warmup_all,
|
|
)
|
|
|
|
FRONTEND_DIR = os.path.join(os.path.dirname(os.path.dirname(__file__)), "frontend")
|
|
|
|
app = FastAPI(title="한국어 TTS 스튜디오", version="2.0.0")
|
|
|
|
app.add_middleware(
|
|
CORSMiddleware,
|
|
allow_origins=["*"],
|
|
allow_methods=["*"],
|
|
allow_headers=["*"],
|
|
)
|
|
|
|
|
|
class TTSRequest(BaseModel):
|
|
text: str = Field(..., description="읽을 텍스트")
|
|
engine: str | None = Field(None, description="엔진 ID (melo / coqui-kss 등)")
|
|
language: str = Field("KR", description="언어 코드")
|
|
speaker: str | None = Field(None, description="화자/모델 ID")
|
|
speed: float = Field(1.4, ge=0.5, le=2.0, description="말하기 속도")
|
|
pitch: float = Field(0.0, ge=-12.0, le=12.0, description="피치(반음)")
|
|
|
|
|
|
@app.on_event("startup")
|
|
def _startup() -> None:
|
|
warmup_all()
|
|
|
|
|
|
@app.get("/api/health")
|
|
def health() -> dict:
|
|
import torch
|
|
|
|
return {
|
|
"status": "ok",
|
|
"cuda": torch.cuda.is_available(),
|
|
"gpu": torch.cuda.get_device_name(0) if torch.cuda.is_available() else None,
|
|
"engines": [e.id for e in list_engines()],
|
|
"default_engine": default_engine_id(),
|
|
}
|
|
|
|
|
|
@app.get("/api/engines")
|
|
def engines() -> dict:
|
|
return {
|
|
"default": default_engine_id(),
|
|
"engines": [e.describe() for e in list_engines()],
|
|
}
|
|
|
|
|
|
@app.post("/api/tts")
|
|
def tts(req: TTSRequest) -> Response:
|
|
engine_id = req.engine or default_engine_id()
|
|
try:
|
|
engine = get_engine(engine_id)
|
|
wav = engine.synth_wav(
|
|
text=req.text,
|
|
language=req.language,
|
|
speaker=req.speaker,
|
|
speed=req.speed,
|
|
pitch=req.pitch,
|
|
)
|
|
except ValueError as exc:
|
|
raise HTTPException(status_code=400, detail=str(exc))
|
|
except Exception as exc: # pragma: no cover
|
|
raise HTTPException(status_code=500, detail=f"합성 실패: {exc}")
|
|
|
|
return Response(
|
|
content=wav,
|
|
media_type="audio/wav",
|
|
headers={"Content-Disposition": 'inline; filename="tts.wav"'},
|
|
)
|
|
|
|
|
|
@app.get("/")
|
|
def index() -> FileResponse:
|
|
return FileResponse(os.path.join(FRONTEND_DIR, "index.html"))
|
|
|
|
|
|
app.mount("/static", StaticFiles(directory=FRONTEND_DIR), name="static")
|