import gradio as gr
from pytube import YouTube
from moviepy.editor import AudioFileClip
from deepgram import DeepgramClient, PrerecordedOptions, FileSource

def download_video(video_url, path='./'):
    yt = YouTube(video_url)
    video = yt.streams.filter(only_audio=True).first()
    out_file = video.download(output_path=path)
    return out_file

def extract_audio(video_path):
    audio_path = video_path.replace(".mp4", ".wav")
    clip = AudioFileClip(video_path)
    clip.write_audiofile(audio_path)
    return audio_path

def transcribe_audio(audio_path, api_key):
    deepgram = DeepgramClient(api_key)
    with open(audio_path, 'rb') as file:
        audio_data = file.read()

    options = PrerecordedOptions(
        model="latest",
        punctuate=True,
        diarize=True
    )

    response = deepgram.transcription.prerecorded(FileSource(file=audio_data), options)
    transcripts = response['results']['channels'][0]['alternatives'][0]['transcript']
    return transcripts

def process_video(video_url, api_key):
    video_path = download_video(video_url)
    audio_path = extract_audio(video_path)
    transcription = transcribe_audio(audio_path, api_key)
    return transcription

# Define Gradio interface
iface = gr.Interface(
    fn=process_video,
    inputs=["text", "text"],  # Accept video URL and API key as inputs
    outputs="text",
    examples=[["https://www.youtube.com/watch?v=jwjjWHzEaJc", "Deepgram API Key"]],

    title="Video to Text Transcription",
    description="Enter a YouTube video URL and your Deepgram API key to extract audio and transcribe it."
)

if __name__ == "__main__":
    iface.launch()