init commit

.vscode/settings.json

@@ -0,0 +1,5 @@
+{
+    "python-envs.defaultEnvManager": "ms-python.python:conda",
+    "python-envs.defaultPackageManager": "ms-python.python:conda",
+    "python-envs.pythonProjects": []
+}

Dockerfile

@@ -0,0 +1,45 @@
+# Start from Ollama's official image
+FROM ollama/ollama
+
+# Remove the default entrypoint
+ENTRYPOINT []
+
+# Install Python essentials
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    python3 \
+    python3-pip \
+    python3-venv \
+    curl \
+    && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+# Set up Virtual Env
+RUN python3 -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Install requirements
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy your code
+COPY . .
+
+# FIX: The user with UID 1000 already exists in this image, 
+# we just need to make sure they own the /app and the ollama path.
+RUN chown -R 1000:1000 /app && \
+    mkdir -p /home/ollama/.ollama && \
+    chown -R 1000:1000 /home/ollama/.ollama
+
+# Set the home and model path for the existing user
+ENV HOME=/home/ollama \
+    OLLAMA_MODELS=/home/ollama/.ollama
+
+# Switch to the existing user (UID 1000)
+USER 1000
+
+# Hugging Face standard port
+EXPOSE 7860
+
+# Startup command
+CMD sh -c "ollama serve & sleep 5 && ollama pull llama3.2:1b && uvicorn main:app --host 0.0.0.0 --port 7860 --timeout-keep-alive 65"

main.py

@@ -0,0 +1,99 @@
+import os
+import json
+import httpx
+import asyncio
+import logging
+from fastapi import FastAPI, HTTPException
+from fastapi.responses import StreamingResponse
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+import edge_tts
+import uvicorn
+import base64
+
+# إعدادات اللوج
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI()
+
+# تفعيل CORS للاتصال مع Next.js
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+OLLAMA_URL = "http://localhost:11434/api/chat"
+
+class ChatRequest(BaseModel):
+    messages: list
+    voice: str = "ar-SA-HamedNeural"
+    rate: str = "+0%"
+
+async def stream_text_and_voice(payload,messages, voice, rate):
+    full_response_text = ""
+    sentence_buffer = ""
+    
+    async with httpx.AsyncClient(timeout=None) as client:
+        try:
+            async with client.stream("POST", OLLAMA_URL, json=payload) as response:
+                async for line in response.aiter_lines():
+                    if not line: continue
+                    chunk = json.loads(line)
+                    token = chunk.get("message", {}).get("content", "")
+                    
+                    sentence_buffer += token
+                    full_response_text += token
+
+                    # Check for sentence end
+                    if any(punct in token for punct in [".", "!", "?", "؟", "\n"]):
+                        clean_text = sentence_buffer.strip()
+                        # print(clean_text)
+                        if clean_text:
+                            # 1. إنشاء كائن التواصل مع edge-tts
+                            communicate = edge_tts.Communicate(clean_text, voice, rate=rate)
+                            
+                            async for chunk in communicate.stream():
+                                if (chunk["type"] == "audio"):
+                                    audio_base64 = base64.b64encode(chunk["data"]).decode('utf-8')
+                                    yield f'{{ "type": "audio", "data": "{audio_base64}" }}\n'
+                                
+                            sentence_buffer = "" # تصغير البفر لبدء جملة جديدة
+
+                # Handle remaining text in buffer
+                if sentence_buffer.strip():
+                    communicate = edge_tts.Communicate(sentence_buffer.strip(), voice, rate=rate)
+                    async for audio_chunk in communicate.stream():
+                        if audio_chunk["type"] == "audio":
+                            b64_data = base64.b64encode(audio_chunk["data"]).decode('utf-8')
+                            yield json.dumps({"type": "audio", "data": b64_data}) + "\n"
+
+            # THE IMPORTANT PART: Send the text message at the end
+            yield json.dumps({
+                "type": "final_text", 
+                "content": full_response_text
+            }) + "\n"
+
+        except Exception as e:
+            logger.error(f"Error: {e}")
+
+@app.post("/stream-voice")
+async def voice_engine(data: ChatRequest):
+    payload = {
+        "model": "llama3.2:1b", # Or whatever model you are using
+        "messages": data.messages,
+        "stream": True,  # Crucial for streaming
+        "options": {
+            "temperature": 0.5,
+            "top_p": 0.9,
+        }
+    }
+    return StreamingResponse(
+        stream_text_and_voice(payload,data.messages, data.voice, data.rate),
+        media_type="audio/mpeg",
+        headers={"Cache-Control":"no-cache"}
+    )
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=7860)

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+edge-tts
+httpx
+pydantic