FROM python:3.11-slim
WORKDIR /app
RUN apt-get update && apt-get install -y --no-install-recommends ffmpeg && \
rm -rf /var/lib/apt/lists/*
RUN pip install --no-cache-dir fastapi uvicorn edge-tts
RUN cat <<'PYEOF' > /app/main.py
from fastapi import FastAPI, Query
from fastapi.responses import StreamingResponse, HTMLResponse
from fastapi.middleware.cors import CORSMiddleware
import edge_tts
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
VOICES = [
"en-US-AvaNeural", "en-US-AndrewNeural", "en-GB-SoniaNeural",
"en-GB-RyanNeural", "ja-JP-NanamiNeural", "ja-JP-KeitaNeural",
"de-DE-KatjaNeural", "fr-FR-DeniseNeural", "es-ES-ElviraNeural",
"zh-CN-XiaoxiaoNeural", "zh-CN-YunjianNeural", "ko-KR-SunHiNeural"
]
@app.get("/")
async def root():
return {"message": "Edge TTS API", "voices": VOICES}
@app.get("/tts")
async def tts(
text: str = Query(..., description="Text to speak"),
voice: str = Query(default="en-US-AvaNeural", description="Voice ID"),
rate: str = Query(default="+0%", description="Speed: +10% or -10%"),
pitch: str = Query(default="+0Hz", description="Pitch: +10Hz or -10Hz")
):
# edge-tts regex validation requires a leading "+" or "-" sign
if not rate.startswith(("+", "-")):
rate = f"+{rate}"
if not pitch.startswith(("+", "-")):
pitch = f"+{pitch}"
communicate = edge_tts.Communicate(text, voice, rate=rate, pitch=pitch)
async def audio_generator():
async for chunk in communicate.stream():
if chunk["type"] == "audio":
yield chunk["data"]
return StreamingResponse(audio_generator(), media_type="audio/mpeg")
@app.get("/ui", response_class=HTMLResponse)
async def ui():
return """
Edge TTS
Edge TTS API
"""
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=7860)
PYEOF
EXPOSE 7860
CMD ["python", "/app/main.py"]