app.py

from fastapi import FastAPI, Request, WebSocket, WebSocketDisconnect
from fastapi.responses import StreamingResponse, JSONResponse
import outetts
import io
import json
# Initialize the interface
interface = outetts.Interface(
    config=outetts.ModelConfig.auto_config(
        model=outetts.Models.VERSION_1_0_SIZE_1B,
        # For llama.cpp backend
        #backend=outetts.Backend.LLAMACPP,
        #quantization=outetts.LlamaCppQuantization.FP16
        # For transformers backend
        backend=outetts.Backend.HF,
    )
)

# Load the default speaker profile
speaker = interface.load_default_speaker("EN-FEMALE-1-NEUTRAL")

app = FastAPI()

@app.get("/")
def greet_json():
    return {"Hello": "World!"}

@app.websocket("/ws/tts")
async def websocket_tts(websocket: WebSocket):
    await websocket.accept()
    try:
        while True:
            # Empfange Text-Chunk vom Client
            data = await websocket.receive_text()
            # Generiere Audio aus Text
            output = interface.generate(
                config=outetts.GenerationConfig(
                    text=data,
                    generation_type=outetts.GenerationType.CHUNKED,
                    speaker=speaker,
                    sampler_config=outetts.SamplerConfig(
                        temperature=0.4
                    ),
                )
            )
             # Speichere Audio temporär als Datei
            temp_path = "temp.wav"
            output.save(temp_path)
            with open(temp_path, "rb") as f:
                audio_bytes = f.read()
            import os
            os.remove(temp_path)
            # Sende Audiodaten als Bytes zurück
            await websocket.send_bytes(audio_bytes)
    except WebSocketDisconnect:
        pass