import warnings; warnings.filterwarnings("ignore")  # This hides warning messages.
import numpy as np
np.random.seed(42)
 

import os  # Allows us to use operating system features.
from openai import OpenAI  # This is the main OpenAI client library.
 

API_key = os.environ["OPENAI_API_KEY"]  # Fetch our API key from the environment.
 

client = OpenAI(api_key=API_key)  # Create the OpenAI client with your key.
 

# Let us try a basic Chat Completion.
response = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)  # Print the response text.

# Use input() to send your own chat prompt.
user_msg = input("Type your message for the assistant: ")
my_response = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[{"role": "user", "content": user_msg}]
)
print(my_response.choices[0].message.content)

# Prepare to use speech recognition.
audio_file = open("sample.wav", "rb")  # Open an audio file. Replace with your file if needed.
transcript = client.audio.transcriptions.create(
    model="whisper-1",
    file=audio_file
)
print(transcript.text)  # Show the transcribed text.

# Try transcribing a different audio file with input().
file_path = input("Enter the path to your audio file: ")
with open(file_path, "rb") as my_audio:
    my_transcription = client.audio.transcriptions.create(
        model="whisper-1",
        file=my_audio
    )
print(my_transcription.text)

# Create an embedding for a simple sentence.
embed = client.embeddings.create(
    model="text-embedding-3-small",
    input=["Machine learning is fun"]
)
print(len(embed.data[0].embedding))  # Shows the vector size.

# Compare two sentences using their embeddings.
import numpy as np  # Lets us do math with arrays.
sentences = ["OpenAI makes AI tools.", "Artificial intelligence tools by OpenAI."]
embs = client.embeddings.create(
    model="text-embedding-3-small",
    input=sentences
)
a, b = np.array(embs.data[0].embedding), np.array(embs.data[1].embedding)
# Use cosine similarity to compare.
similarity = np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))
print("Similarity:", similarity)

# Wrap a chat completion call with error handling.
try:
    resp = client.chat.completions.create(
        model="gpt-4.1-mini",
        messages=[{"role": "user", "content": "Give me a joke."}]
    )
    print(resp.choices[0].message.content)
except Exception as e:
    print("Error: ", e)

# Mini Project: Turn an audio file to text and summarize it using chat.
audio_file = open("sample.wav", "rb")
transcript = client.audio.transcriptions.create(
    model="whisper-1",
    file=audio_file
)
summary = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[
        {"role": "system", "content": "You summarize transcripts."},
        {"role": "user", "content": transcript.text}
    ]
)
print("Summary:\n", summary.choices[0].message.content)

# Challenge 1: Try summarizing your own audio file.
file_path = input("Type the path to your own audio file: ")
with open(file_path, "rb") as my_audio:
    my_transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=my_audio
    )
    my_summary = client.chat.completions.create(
        model="gpt-4.1-mini",
        messages=[
            {"role": "system", "content": "You summarize transcripts."},
            {"role": "user", "content": my_transcript.text}
        ]
    )
    print("Summary:\n", my_summary.choices[0].message.content)

# Challenge 2: Get keyword ideas using chat completions.
topic = input("Enter a topic or theme: ")
ideas = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[
        {"role": "system", "content": "You produce a list of keywords."},
        {"role": "user", "content": topic}
    ]
)
print("Keywords:\n", ideas.choices[0].message.content)