app.py

from transformers import pipeline
import gradio as gr
import moviepy.editor as mp
from pytube import YouTube

pipe = pipeline(model="Campfireman/whisper-small-hi")  # change to "your-username/the-name-you-picked"

def download_video(url):
    print("Downloading...")
    local_file = (
        YouTube(url)
        .streams.filter(progressive=True, file_extension="mp4")
        .first()
        .download()
    )
    print("Downloaded")
    return local_file

def validate_youtube(url):
    #This creates a youtube object
    try:
        yt = YouTube(url)  
    except Exception:
        print("Hi there URL seems invalid")
        return True
    #This will return the length of the video in sec as an int
    video_length = yt.length
    if video_length > 600:
        print("Your video is larger than 10 minutes")
        return True
    else:
        print("Your video is less than 10 minutes")
        return False

def validate_url(url):
    import validators
    if not validators.url(url):
        return True
    else:
        return False  

def download_video(url):
    print("Downloading...")
    local_file = (
        YouTube(url)
        .streams.filter(progressive=True, file_extension="mp4")
        .first()
        .download()
    )
    print("Downloaded")
    return local_file

def video_spliter(video_in):
    my_clip = mp.VideoFileClip(download_video(video_in))
    my_audio = "audio_out.wav"
    my_clip.audio.write_audiofile(my_audio)
    return my_audio

def transcribe(video_url):
    if validate_url(video_url):
        if not validate_youtube(video_url):
            return "The URL seems not for Youtube videos"
    else: 
        return "The URL seems invalid"
    
    audio = video_spliter(video)
    
    #text = pipe(audio)["text"]
    return text



iface = gr.Interface(
    fn=transcribe, 
    inputs=gr.Textbox(label = "Enter the URL of the Youtube video clip here: "), 
    outputs="text",
    title="Whisper Medium Chinese",
    description="Video Chinese Transcriptior",
)

iface.launch()