Create app.py

app.py

@@ -0,0 +1,108 @@
+import gradio as gr
+import requests
+import base64
+from io import BytesIO
+from PIL import Image
+import os
+
+# 🔐 Use environment variable for your key (set it in Hugging Face "Secrets")
+API_KEY = os.getenv("GOOGLE_API_KEY")
+
+# ---------- Helper Functions ---------- #
+
+def text_generation(prompt):
+    url = "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent"
+    headers = {"Content-Type": "application/json", "x-goog-api-key": API_KEY}
+    data = {"contents": [{"parts": [{"text": prompt}]}]}
+    res = requests.post(url, headers=headers, json=data)
+    output = res.json()
+    try:
+        return output["candidates"][0]["content"]["parts"][0]["text"]
+    except Exception:
+        return str(output)
+
+
+def image_generation(prompt):
+    url = "https://generativelanguage.googleapis.com/v1beta/models/imagen-3.0-generate:generateImage"
+    headers = {"Content-Type": "application/json", "x-goog-api-key": API_KEY}
+    data = {"prompt": {"text": prompt}}
+    res = requests.post(url, headers=headers, json=data)
+    result = res.json()
+    try:
+        img_b64 = result["images"][0]["data"]
+        image_bytes = base64.b64decode(img_b64)
+        return Image.open(BytesIO(image_bytes))
+    except Exception:
+        return str(result)
+
+
+def video_generation(prompt):
+    url = "https://generativelanguage.googleapis.com/v1beta/models/veo-2.0-generate-001:generateVideo"
+    headers = {"Content-Type": "application/json", "x-goog-api-key": API_KEY}
+    data = {"prompt": {"text": prompt}}
+    res = requests.post(url, headers=headers, json=data)
+    result = res.json()
+    try:
+        video_url = result["videos"][0].get("url", None)
+        if video_url:
+            return video_url
+        else:
+            return "Video generated (but no URL returned): " + str(result)
+    except Exception:
+        return str(result)
+
+
+def text_to_speech(text):
+    url = "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-tts:generateContent"
+    headers = {"Content-Type": "application/json", "x-goog-api-key": API_KEY}
+    data = {"contents": [{"parts": [{"text": text}]}]}
+    res = requests.post(url, headers=headers, json=data)
+    result = res.json()
+    try:
+        audio_b64 = result["candidates"][0]["content"]["parts"][0]["inline_data"]["data"]
+        audio_bytes = base64.b64decode(audio_b64)
+        return (audio_bytes, "audio/wav")
+    except Exception:
+        return str(result)
+
+# ---------- Gradio Interfaces ---------- #
+
+text_tab = gr.Interface(
+    fn=text_generation,
+    inputs=gr.Textbox(label="Enter your text prompt"),
+    outputs=gr.Textbox(label="Generated Text"),
+    title="🧠 Text Generation (Gemini 2.5 Flash)"
+)
+
+image_tab = gr.Interface(
+    fn=image_generation,
+    inputs=gr.Textbox(label="Enter image prompt"),
+    outputs=gr.Image(type="pil"),
+    title="🖼️ Image Generation (Imagen 3)"
+)
+
+video_tab = gr.Interface(
+    fn=video_generation,
+    inputs=gr.Textbox(label="Describe your video idea"),
+    outputs=gr.Textbox(label="Generated Video URL / Response"),
+    title="🎥 Video Generation (Veo 2)"
+)
+
+tts_tab = gr.Interface(
+    fn=text_to_speech,
+    inputs=gr.Textbox(label="Enter text for speech"),
+    outputs=gr.Audio(label="Generated Speech"),
+    title="🗣️ Text-to-Speech (Gemini TTS)"
+)
+
+# ---------- Combine All Tabs ---------- #
+
+demo = gr.TabbedInterface(
+    [text_tab, image_tab, video_tab, tts_tab],
+    tab_names=["Text", "Image", "Video", "Speech"]
+)
+
+# ---------- Launch ---------- #
+
+if __name__ == "__main__":
+    demo.launch()