{
"cells": [
{
"cell_type": "markdown",
"id": "ad6e5ed4-a38d-46a6-8bb5-32d68bd0b9e5",
"metadata": {},
"source": [
"End of week 2 exercise"
]
},
{
"cell_type": "code",
"execution_count": 38,
"id": "8f45fbfa-eaaa-4eb8-841e-83b068b80507",
"metadata": {},
"outputs": [],
"source": [
"# imports\n",
"\n",
"import os\n",
"import json\n",
"import gradio as gr\n",
"import base64\n",
"from dotenv import load_dotenv\n",
"from openai import OpenAI\n",
"from io import BytesIO\n",
"from PIL import Image\n",
"from pydub import AudioSegment\n",
"from pydub.playback import play\n"
]
},
{
"cell_type": "code",
"execution_count": 39,
"id": "451b72a6-1e6c-476a-8431-1c30c5cd9fb8",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"API key found and looks good so far!\n"
]
}
],
"source": [
"# Initialization\n",
"\n",
"load_dotenv(override=True)\n",
"openai_api_key = os.getenv('OPENAI_API_KEY')\n",
"if openai_api_key:\n",
" print(\"API key found and looks good so far!\")\n",
"else:\n",
" print(\"No API key was found!\")\n",
"\n",
"MODEL = \"gpt-4o-mini\"\n",
"openai = OpenAI()"
]
},
{
"cell_type": "code",
"execution_count": 40,
"id": "29fa6a53-4b57-47ea-89a1-640020e603b4",
"metadata": {},
"outputs": [],
"source": [
"system_message = (\n",
" \"You are a helpful tennis coach who answers questions about tennis rules, \"\n",
" \"players, strategies, training, and equipment.\"\n",
" \"Give short, courteous answers, no more than 2 sentence.\"\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 41,
"id": "e9b255e7-02d8-4350-b5d4-e645d1fc90d3",
"metadata": {},
"outputs": [],
"source": [
"# Translation\n",
"\n",
"LANG_CODES = {\n",
" \"English\": \"en\",\n",
" \"Spanish\": \"es\",\n",
" \"French\": \"fr\"\n",
"}\n",
"\n",
"def translate_text(text, target_language=\"en\"):\n",
" messages = [\n",
" {\"role\": \"system\", \"content\": f\"You are a translator. Translate the following text to {target_language}\"},\n",
" {\"role\": \"user\", \"content\": text}\n",
" ]\n",
" response = openai.chat.completions.create(model=MODEL, messages=messages)\n",
" return response.choices[0].message.content\n",
"\n",
"def tennis_info_tool(query):\n",
" if \"top\" in query.lower():\n",
" return \"Top male players: Djokovic, Nadal, Federer. Top female players: Barty, Sabalenka, Swiatek.\"\n"
]
},
{
"cell_type": "code",
"execution_count": 42,
"id": "b44b147c-bfba-4137-9ecb-d5538f08a46d",
"metadata": {},
"outputs": [],
"source": [
"# Image\n",
"def generate_tennis_image(prompt):\n",
" image_response = openai.images.generate(\n",
" model=\"dall-e-3\",\n",
" prompt=f\"Tennis scene: {prompt}, realistic and detailed, vibrant colors\",\n",
" size=\"1024x1024\",\n",
" n=1,\n",
" response_format=\"b64_json\",\n",
" )\n",
" image_base64 = image_response.data[0].b64_json\n",
" image_data = base64.b64decode(image_base64)\n",
" return Image.open(BytesIO(image_data))"
]
},
{
"cell_type": "code",
"execution_count": 43,
"id": "bbfeff3b-0c73-4b2c-a6da-3cac27d8fedd",
"metadata": {},
"outputs": [],
"source": [
"# Audio\n",
"\n",
"def talker(message):\n",
" response = openai.audio.speech.create(\n",
" model=\"tts-1\",\n",
" voice=\"onyx\",\n",
" input=message\n",
" )\n",
"\n",
" audio_stream = BytesIO(response.content)\n",
" audio = AudioSegment.from_file(audio_stream, format=\"mp3\")\n",
" play(audio)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "fa338332-3dfc-4b95-8367-65853a8d2793",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 44,
"id": "574d672e-0a75-4af9-b3ad-8dc2dec4e607",
"metadata": {},
"outputs": [],
"source": [
"def chat(history, user_message, target_language=\"English\", use_audio=False, generate_image=False):\n",
" image = None\n",
"\n",
" if any(keyword in user_message.lower() for keyword in [\"top\", \"players\"]):\n",
" reply = tennis_info_tool(user_message)\n",
" else:\n",
" messages = [{\"role\": \"system\", \"content\": system_message}] + history\n",
" response = openai.chat.completions.create(model=MODEL, messages=messages)\n",
" reply = response.choices[0].message.content\n",
" \n",
" if target_language != \"English\":\n",
" code = LANG_CODES.get(target_language, \"en\")\n",
" reply = translate_text(reply, code)\n",
"\n",
" history.append({\"role\": \"assistant\", \"content\": reply})\n",
" \n",
" if use_audio:\n",
" talker(reply)\n",
"\n",
" if generate_image:\n",
" image = generate_tennis_image(reply)\n",
" return history, image"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "eaf4f47e-d20b-41f8-94b5-4aef0302731b",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 45,
"id": "9110343f-0efa-49bc-8d5f-498fd690dd14",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"* Running on local URL: http://127.0.0.1:7869\n",
"* To create a public link, set `share=True` in `launch()`.\n"
]
},
{
"data": {
"text/html": [
""
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": []
},
"execution_count": 45,
"metadata": {},
"output_type": "execute_result"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Input #0, wav, from '/var/folders/73/0s09hh2n48q7s14tld64q3rh0000gn/T/tmp4hoe_x5n.wav':\n",
" Duration: 00:00:06.55, bitrate: 384 kb/s\n",
" Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 24000 Hz, 1 channels, s16, 384 kb/s\n",
" 6.45 M-A: -0.000 fd= 0 aq= 0KB vq= 0KB sq= 0B "
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Input #0, wav, from '/var/folders/73/0s09hh2n48q7s14tld64q3rh0000gn/T/tmp2mxw0wth.wav':\n",
" Duration: 00:00:04.61, bitrate: 384 kb/s\n",
" Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 24000 Hz, 1 channels, s16, 384 kb/s\n",
" 4.48 M-A: 0.000 fd= 0 aq= 0KB vq= 0KB sq= 0B "
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"# Gradio\n",
"\n",
"with gr.Blocks() as ui:\n",
" with gr.Row():\n",
" chatbot = gr.Chatbot(height=500, type=\"messages\")\n",
" image_output = gr.Image(height=500)\n",
" with gr.Row():\n",
" entry = gr.Textbox(label=\"Ask your Tennis Coach:\", placeholder=\"Type and press Enter...\")\n",
" target_lang_dropdown = gr.Dropdown(\n",
" choices=[\"English\", \"Spanish\", \"French\"],\n",
" value=\"English\",\n",
" label=\"Translate to:\"\n",
" )\n",
" audio_toggle = gr.Checkbox(label=\"Play audio\", value=False)\n",
" image_toggle = gr.Checkbox(label=\"Generate image for this reply\", value=True)\n",
" with gr.Row():\n",
" clear = gr.Button(\"Clear\")\n",
" \n",
" def add_user_message(message, history):\n",
" history.append({\"role\": \"user\", \"content\": message})\n",
" return \"\", history\n",
"\n",
" def chat_response(history, message, target_language, use_audio, generate_image):\n",
" history, image = chat(history, message, target_language, use_audio, generate_image)\n",
" return history, image\n",
"\n",
" entry.submit(\n",
" add_user_message,\n",
" inputs=[entry, chatbot],\n",
" outputs=[entry, chatbot]).then(\n",
" chat_response,\n",
" inputs=[chatbot, entry, target_lang_dropdown, audio_toggle, image_toggle],\n",
" outputs=[chatbot, image_output]\n",
" )\n",
"\n",
" clear.click(lambda: (None, None, None), inputs=None, outputs=[chatbot, image_output, entry], queue=False)\n",
"\n",
"ui.launch(inbrowser=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "f0359c29-22aa-4156-9afa-8c63c02ca747",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.13"
}
},
"nbformat": 4,
"nbformat_minor": 5
}