Update app.py

app.py

@@ -1,22 +1,26 @@
 from fastapi import FastAPI, Form
 from fastapi.responses import FileResponse, JSONResponse
 from pydantic import BaseModel
-from google import generativeai as genai
-from google.generativeai import types
+import google.generativeai as genai
+from google.generativeai.types import GenerateContentRequest, GenerateContentResponse, Content, Part
+from google.generativeai.types import GenerateContentResponse, GenerateContentRequest, GenerationConfig
+from google.generativeai.types import SafetySetting, HarmBlockThreshold, HarmCategory, GenerateContentRequest, Content
 import wave
 import os
 
 app = FastAPI()
 
-# Set your Google API key via environment variable for Spaces safety
+# Set your Google API key via environment variable
 GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
 
 if not GOOGLE_API_KEY:
     raise ValueError("GOOGLE_API_KEY environment variable not set.")
 
-# Initialize the GenAI client
-client = genai.Client(api_key=GOOGLE_API_KEY)
+# Configure the GenAI client
+genai.configure(api_key=GOOGLE_API_KEY)
 
+# Load the TTS model
+model = genai.GenerativeModel(model_name="gemini-2.5-flash-preview-tts")
 
 # Function to write PCM audio data to WAV file
 def wave_file(filename, pcm, channels=1, rate=24000, sample_width=2):
@@ -27,32 +31,36 @@ def wave_file(filename, pcm, channels=1, rate=24000, sample_width=2):
         wf.writeframes(pcm)
 
 
-# Root endpoint to check API is running
+# Root endpoint to check API status
 @app.get("/")
 def read_root():
-    return {"message": "🦜 Gemini TTS FastAPI running on Hugging Face Spaces!"}
+    return {"message": "✅ Gemini TTS FastAPI running on Hugging Face Spaces!"}
 
 
 # POST endpoint to generate TTS audio from text
 @app.post("/generate-audio")
 def generate_audio(text: str = Form(...)):
     try:
-        response = client.models.generate_content(
-            model="gemini-2.5-flash-preview-tts",
+        response = model.generate_content(
             contents=text,
-            config=types.GenerateContentConfig(
-                response_modalities=["AUDIO"],
-                speech_config=types.SpeechConfig(
-                    voice_config=types.VoiceConfig(
-                        prebuilt_voice_config=types.PrebuiltVoiceConfig(
-                            voice_name="Kore"
-                        )
+            generation_config=genai.types.GenerationConfig(
+                response_mime_type="audio/wav"
+            ),
+            safety_settings=[
+                SafetySetting(category=HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT, threshold=HarmBlockThreshold.BLOCK_NONE)
+            ],
+            tools=None,
+            response_modality="AUDIO",
+            speech_config=genai.types.SpeechConfig(
+                voice_config=genai.types.VoiceConfig(
+                    prebuilt_voice_config=genai.types.PrebuiltVoiceConfig(
+                        voice_name="Kore"
                     )
-                ),
+                )
             )
         )
 
-        # Get audio data
+        # Get raw audio data
         data = response.candidates[0].content.parts[0].inline_data.data
 
         # Save to file