app.py

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

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):
    os.environ["DEEPGRAM_API_KEY"] = api_key
    # Environment Setup (Ensure environment variables are set appropriately)
    load_dotenv()
    # Deepgram API Key Setup
    deepgram_api_key = os.getenv('DEEPGRAM_API_KEY')  # Ensure DG_KEY is set in your environment variables
    deepgram = DeepgramClient("", ClientOptionsFromEnv())
    with open(audio_path, 'rb') as file:
        audio_data = file.read()

    payload: FileSource = {
            "buffer": audio_data,
        }

    options = PrerecordedOptions(
            model="nova-2",
            smart_format=True,
            utterances=True,
            punctuate=True,
            diarize=True,
            detect_language =True,
            sentiment=True,
            summarize=True,
            topics=True,
            intents=True
        )

    response =  deepgram.listen.prerecorded.v("1").transcribe_file(payload, options)

    #response_buffer =  deepgram.listen.prerecorded(payload, options)


        # STEP 3: Call the analyze_text method with the text payload and options

    #print(response.to_json(indent=4))
    print("")
    transcripts = response['results']['channels'][0]['alternatives'][0]['transcript'][0]

    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()