From e7700cf6e65fd737bf51962f8a5c5cb2414d5ab6 Mon Sep 17 00:00:00 2001
From: Lukas Holzner <git@lukas.lhal.de>
Date: Tue, 1 Oct 2024 11:10:47 +0200
Subject: [PATCH] Added tts endpoint

---
 backend/app/ai_manager.py  | 36 +++++++++++++++++++++++++++++++++++-
 backend/app/main.py        |  3 ++-
 backend/app/routers/tts.py | 10 ++++++++++
 3 files changed, 47 insertions(+), 2 deletions(-)
 create mode 100644 backend/app/routers/tts.py

diff --git a/backend/app/ai_manager.py b/backend/app/ai_manager.py
index 8328fda..eba4410 100644
--- a/backend/app/ai_manager.py
+++ b/backend/app/ai_manager.py
@@ -1,10 +1,13 @@
 import os
 from dotenv import load_dotenv
-from fastapi import UploadFile
+from fastapi import UploadFile, HTTPException
+from fastapi.responses import FileResponse
 from openai import OpenAI
 from typing import List, Dict
 import json
 import io
+import tempfile
+from starlette.background import BackgroundTask
 
 # Load environment variables from .env file
 load_dotenv()
@@ -160,3 +163,34 @@ async def convert_audio_to_text(audio_file: UploadFile):
     except Exception as e:
         print(f"Error in audio transcription: {str(e)}")
         return None
+
+async def convert_text_to_audio(text: str, voice: str = "alloy"):
+    """
+    Convert text to audio using OpenAI's text-to-speech API and return a FastAPI FileResponse.
+    
+    :param text: The text to convert to speech
+    :param voice: The voice to use (default is "alloy")
+    :return: A FastAPI FileResponse containing the audio file
+    """
+    try:
+        response = client.audio.speech.create(
+            model="tts-1",
+            voice=voice,
+            input=text
+        )
+        
+        # Create a temporary file to store the audio content
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_file:
+            temp_file.write(response.content)
+            temp_file_path = temp_file.name
+
+        # Return a FileResponse
+        return FileResponse(
+            path=temp_file_path,
+            media_type="audio/mpeg",
+            filename="speech.mp3",
+            background=BackgroundTask(lambda: os.unlink(temp_file_path))
+        )
+    except Exception as e:
+        print(f"Error in text-to-speech conversion: {str(e)}")
+        raise HTTPException(status_code=500, detail="Text-to-speech conversion failed")
diff --git a/backend/app/main.py b/backend/app/main.py
index 3b71314..a8ae498 100644
--- a/backend/app/main.py
+++ b/backend/app/main.py
@@ -6,7 +6,7 @@
 from fastapi.middleware.cors import CORSMiddleware
 from .database import engine
 from . import models
-from .routers import meetings, users
+from .routers import meetings, users, tts
 import os
 
 models.Base.metadata.create_all(bind=engine)
@@ -34,6 +34,7 @@
 # Include the router with the /api prefix
 app.include_router(meetings.router, tags=["Meetings"])
 app.include_router(users.router, tags=["Users"])
+app.include_router(tts.router, tags=["TTS"])
 
 @app.get("/", include_in_schema=False)
 async def serve_index():
diff --git a/backend/app/routers/tts.py b/backend/app/routers/tts.py
new file mode 100644
index 0000000..ba2fa66
--- /dev/null
+++ b/backend/app/routers/tts.py
@@ -0,0 +1,10 @@
+from fastapi import APIRouter, Body
+from fastapi.responses import FileResponse
+from app.ai_manager import convert_text_to_audio
+
+
+router = APIRouter()
+
+@router.post("/tts/")
+async def tts(text: str = Body(...), voice: str = Body("alloy")):
+    return await convert_text_to_audio(text, voice)
\ No newline at end of file